Datasets:
Brendan
/
openhands_ce_data_dev

Dataset card Data Studio Files
xet
 Community
instance_id
stringlengths
55
127
patch
stringclasses
63 values
FAIL_TO_PASS
listlengths
1
4.14k
⌀
PASS_TO_PASS
listlengths
5
5.44k
⌀
image_name
stringclasses
26 values
repo
stringclasses
26 values
problem_statement
stringlengths
261
2.28k
task_id
stringlengths
28
69
domain
stringclasses
8 values
system_prompt
stringclasses
34 values
user_prompt
stringclasses
64 values
selected_tools
listlengths
4
17
⌀
restricted_tools
listlengths
0
0
⌀
mcp_endpoint
stringclasses
1 value
number_of_runs
int64
1
1
⌀
reset_database_between_runs
bool
1 class
gym_servers_config
stringclasses
50 values
verifiers
stringclasses
64 values
benchmark
stringclasses
2 values
enterprise_ops_test_result
stringlengths
503
14.5k
⌀
skillsbench_test_result
stringclasses
0 values
instruction
stringlengths
261
6.16k
num_actions
int64
1
117
conversation_archive_path
stringlengths
180
272
benchmark_run_name
stringclasses
6 values
resolved
bool
2 classes
model_pred_patch
stringlengths
0
70.9k
⌀
model
stringclasses
3 values
eval_report
stringlengths
44
485k
⌀
eval_test_output
stringlengths
6.17k
5.32M
⌀
prompt_tokens
int64
5.78k
11.5M
reasoning_tokens
int64
337
197k
completion_tokens
int64
759
220k
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20251226_121423_362_c4ad7a70_4be55408
null
null
null
null
null
I just received an email from Albert Rodriguez explaining that he needs to take a week of emergency leave due to a sudden family medical situation that requires him to be physically present. Since this type of request affects his time-off balance and may also impact certain benefits tied to his stage in the employee li...
task_20251226_121423_362_c4ad7a70_4be55408
hr
# HR MANAGEMENT ASSISTANT POLICY **Role:** HR Management Assistant **System Scope:** Internal Employee Services, HR Case Management, Personnel Data, and Policy Fulfillment. **Compliance Level:** Strict, especially regarding PII/PHI. You are an HR Management Assistant. Your primary goal is to facilitate the efficien...
I just received an email from Albert Rodriguez explaining that he needs to take a week of emergency leave due to a sudden family medical situation that requires him to be physically present. Since this type of request affects his time-off balance and may also impact certain benefits tied to his stage in the employee li...
[ "link_knowledge_to_hr_case", "send_notification", "create_new_hr_case", "get_user_using_name", "get_hr_service_by_name", "retrieve_knowledge_articles" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-hr-internal", "mcp_server_url": "http://localhost:8008", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hr/dbs/db_1765227945688_occ0a7ed3.sql", "context": {"x-hr-user-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJVU0VSXzAzMSIsImlzcyI6Iml0c20ifQ.Kl0_dflKxwRJSMeKKF2QT4fwpM...
[{"verifier_type": "database_state", "name": "Case Created", "description": "Confirms that the Leave of Absence case was successfully created.", "gym_name": "sn-hr-internal", "validation_config": {"query": "SELECT COUNT(*) FROM hr_case WHERE hr_case_id = 61 AND hr_service = 6 AND opened_for = 92 AND priority = 'moderat...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hr","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":6,"verification_results":{"Case Created":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 1","expected":1,"passed":false,"query":"SELECT COUNT(*) FROM hr_case WHERE hr_case_id = 61...
null
I just received an email from Albert Rodriguez explaining that he needs to take a week of emergency leave due to a sudden family medical situation that requires him to be physically present. Since this type of request affects his time-off balance and may also impact certain benefits tied to his stage in the employee li...
8
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hr_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251226_121423_362_c4ad7a70_4be55408.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
33,350
1,011
2,105
eog-gemini-gemini-3.5-flash-oh-task_20260115_121701_840_1628b966_ff394d2f
null
null
null
null
null
Firstly check for a Draft, if any present, del it. Check if any label tag present for project updates if not simply create it with name "Project Updates" with background color as #ffad46. Also, create a new case ,"Account access issue reported via email - Unable to log in and needs assistance" with priority 'moderate' ...
task_20260115_121701_840_1628b966_ff394d2f
hybrid
# **Email Management Assistant Policy** **Role:** Email Management Assistant **Mandate:** Secure, accurate, and efficient management of user email communications (Messages, Drafts, Threads, Labels, Filters, and Settings). You must act solely based on **confirmed user context** and **existing data integrity rules**. ...
Firstly check for a Draft, if any present, del it. Check if any label tag present for project updates if not simply create it with name "Project Updates" with background color as #ffad46. Also, create a new case ,"Account access issue reported via email - Unable to log in and needs assistance" with priority 'moderate' ...
[ "delete_draft", "create_new_case", "create_label", "list_drafts", "list_labels" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hybrid/dbs/db_1765361971944_6ld5uejbx.sql", "context": {"x-email-user-token": "Kx5XwZK6imPxgyAE88USnS4lF3np3br81Tf9FZr4Vno", "user_id": "104daf4b-2023-49f1-8def-39b4e9dcda5e", "pr...
[{"verifier_type": "database_state", "name": "Verify draft deletion", "description": "Verify draft deletion", "gym_name": "gym-email-mcp", "validation_config": {"query": "SELECT COUNT(*) AS cnt\nFROM drafts d\nJOIN messages m ON m.id = d.message_id\nWHERE d.id = '6b79fe10-757a-4fb1-bde3-56e5adde5987'\n AND m.user_id =...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hybrid","passed_verifiers":3,"resolved":true,"score":1,"total_verifiers":3,"verification_results":{"Verify draft deletion":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 0","expected":0,"passed":true,"query":"SELECT COUNT(*) AS cnt\nFROM drafts d\n...
null
Firstly check for a Draft, if any present, del it. Check if any label tag present for project updates if not simply create it with name "Project Updates" with background color as #ffad46. Also, create a new case ,"Account access issue reported via email - Unable to log in and needs assistance" with priority 'moderate' ...
6
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hybrid_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260115_121701_840_1628b966_ff394d2f.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gemini/gemini-3.5-flash
null
null
56,016
1,995
2,539
eog-gpt-5.5-2026-04-23-oh-task_20260109_074002_906_060e2c81_e0a9e99d
null
null
null
null
null
Locate the latest non-trashed file contains 'Budget'. List its revision history so I can see every saved version. Fetch the most recent revision’s metadata (who changed it, when, and size), and mark that revision to keep forever so it can’t be pruned. Then export this frozen revision as a PDF for audit records and add ...
task_20260109_074002_906_060e2c81_e0a9e99d
drive
# **Drive Management Assistant Policy** **Role:** Drive Management Assistant **Mandate:** Secure and accurate management of user content, permissions, and organization within Google Drive and Shared Drives. You must operate exclusively based on **confirmed user permissions**, **existing file states**, and **database...
Locate the latest non-trashed file contains 'Budget'. List its revision history so I can see every saved version. Fetch the most recent revision’s metadata (who changed it, when, and size), and mark that revision to keep forever so it can’t be pruned. Then export this frozen revision as a PDF for audit records and add ...
[ "get_revisions", "list_files", "update_revisions", "list_revisions", "update_files", "export_files" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-google-drive-mcp", "mcp_server_url": "http://localhost:8009", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/drive/dbs/db_1764776317015_jplw4hf1v.sql", "context": {"X-Auth-Token": "ya29.JYFh8MFYw3YZG9AI8AZyFL3NhGR3QWTE_3DEf7tfuylMyHGDeZHrhq4bXs0puqbBbC8YMtimpmDcmEurRJ0IYYQYwVnueud...
[{"verifier_type": "database_state", "name": "Correct File Selected", "description": "Verify that the agent selected the most recently modified, non-trashed file named 'Project Phoenix Spec'.", "gym_name": "gym-google-drive-mcp", "validation_config": {"query": "WITH candidates AS (SELECT id FROM files WHERE name like '...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"drive","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":3,"verification_results":{"Correct File Selected":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"WITH candidates AS (SELECT id FROM file...
null
Locate the latest non-trashed file contains 'Budget'. List its revision history so I can see every saved version. Fetch the most recent revision’s metadata (who changed it, when, and size), and mark that revision to keep forever so it can’t be pruned. Then export this frozen revision as a PDF for audit records and add ...
7
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-drive_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260109_074002_906_060e2c81_e0a9e99d.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
51,261
2,457
3,216
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-pydantic__pydantic.acb0f10f.func_pm_remove_cond__yd4ldncj
diff --git a/pydantic/json_schema.py b/pydantic/json_schema.py index ca031ea..b8c2330 100644 --- a/pydantic/json_schema.py +++ b/pydantic/json_schema.py @@ -470,21 +470,6 @@ class GenerateJsonSchema: """ # Generate the core-schema-type-specific bits of the schema generation: json_...
[ "tests/test_docs.py::test_docstrings_examples[pydantic/json_schema.py:2586-2620]", "tests/test_json_schema.py::test_byte_size_type", "tests/test_json_schema.py::test_ref_conflict_resolution_with_mode_difference", "tests/test_json_schema.py::test_conflicting_names", "tests/test_json_schema.py::test_plain_ser...
[ "tests/benchmarks/test_attribute_access.py::test_setattr", "tests/benchmarks/test_attribute_access.py::test_getattr", "tests/benchmarks/test_discriminated_unions.py::test_schema_build", "tests/benchmarks/test_discriminated_unions.py::test_efficiency_with_highly_nested_examples", "tests/benchmarks/test_fasta...
swebench/swesmith.x86_64.pydantic_1776_pydantic.acb0f10f
swesmith/pydantic__pydantic.acb0f10f
JSON Schema generation broken for serialization mode with PlainSerializer #### Description When generating JSON schemas in serialization mode, models with `PlainSerializer` annotations are not producing the expected schema output. The serialization schema generation appears to be completely skipped, causing the schem...
pydantic__pydantic.acb0f10f.func_pm_remove_cond__yd4ldncj
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pydantic__pydantic.acb0f10f/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> JSON Schema generation broken for serialization mode with PlainSerializer #### Description When gener...
53
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pydantic__pydantic.acb0f10f.func_pm_remove_cond__yd4ldncj.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/pydantic/json_schema.py b/pydantic/json_schema.py index b8c2330..660171a 100644 --- a/pydantic/json_schema.py +++ b/pydantic/json_schema.py @@ -470,6 +470,21 @@ class GenerateJsonSchema: """ # Generate the core-schema-type-specific bits of the schema generation: json_...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_json_schema.py::test_byte_size_type", "tests/test_json_schema.py::test_ref_conflict_resolution_with_mode_difference", "tests/test_json_s...
+ cd /testbed + : '>>>>> Start Test Output' + /root/.local/bin/uv run pytest --disable-warnings --color=no --tb=no --verbose Building pydantic @ file:///testbed Built pydantic @ file:///testbed Uninstalled 1 package in 1ms Installed 1 package in 7ms ============================= test session starts ===========...
1,433,472
5,534
13,510
eog-gpt-5.5-2026-04-23-oh-task_20251226_121423_362_c4ad7a70_4be55408
null
null
null
null
null
I just received an email from Albert Rodriguez explaining that he needs to take a week of emergency leave due to a sudden family medical situation that requires him to be physically present. Since this type of request affects his time-off balance and may also impact certain benefits tied to his stage in the employee li...
task_20251226_121423_362_c4ad7a70_4be55408
hr
# HR MANAGEMENT ASSISTANT POLICY **Role:** HR Management Assistant **System Scope:** Internal Employee Services, HR Case Management, Personnel Data, and Policy Fulfillment. **Compliance Level:** Strict, especially regarding PII/PHI. You are an HR Management Assistant. Your primary goal is to facilitate the efficien...
I just received an email from Albert Rodriguez explaining that he needs to take a week of emergency leave due to a sudden family medical situation that requires him to be physically present. Since this type of request affects his time-off balance and may also impact certain benefits tied to his stage in the employee li...
[ "link_knowledge_to_hr_case", "send_notification", "create_new_hr_case", "get_user_using_name", "get_hr_service_by_name", "retrieve_knowledge_articles" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-hr-internal", "mcp_server_url": "http://localhost:8008", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hr/dbs/db_1765227945688_occ0a7ed3.sql", "context": {"x-hr-user-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJVU0VSXzAzMSIsImlzcyI6Iml0c20ifQ.Kl0_dflKxwRJSMeKKF2QT4fwpM...
[{"verifier_type": "database_state", "name": "Case Created", "description": "Confirms that the Leave of Absence case was successfully created.", "gym_name": "sn-hr-internal", "validation_config": {"query": "SELECT COUNT(*) FROM hr_case WHERE hr_case_id = 61 AND hr_service = 6 AND opened_for = 92 AND priority = 'moderat...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hr","passed_verifiers":6,"resolved":true,"score":1,"total_verifiers":6,"verification_results":{"Case Created":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) FROM hr_case WHERE hr_case_id = 61 A...
null
I just received an email from Albert Rodriguez explaining that he needs to take a week of emergency leave due to a sudden family medical situation that requires him to be physically present. Since this type of request affects his time-off balance and may also impact certain benefits tied to his stage in the employee li...
12
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hr_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251226_121423_362_c4ad7a70_4be55408.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gpt-5.5-2026-04-23
null
null
43,921
2,043
2,680
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-pydantic__pydantic.acb0f10f.lm_rewrite__y6a8rjdp
diff --git a/pydantic/_internal/_discriminated_union.py b/pydantic/_internal/_discriminated_union.py index 5dd6fda..9daca32 100644 --- a/pydantic/_internal/_discriminated_union.py +++ b/pydantic/_internal/_discriminated_union.py @@ -460,20 +460,16 @@ class _ApplyInferredDiscriminator: code='discriminat...
[ "tests/test_discriminated_union.py::test_distinct_choices" ]
[ "tests/benchmarks/test_attribute_access.py::test_setattr", "tests/benchmarks/test_attribute_access.py::test_getattr", "tests/benchmarks/test_discriminated_unions.py::test_schema_build", "tests/benchmarks/test_discriminated_unions.py::test_efficiency_with_highly_nested_examples", "tests/benchmarks/test_fasta...
swebench/swesmith.x86_64.pydantic_1776_pydantic.acb0f10f
swesmith/pydantic__pydantic.acb0f10f
Discriminator value mapping error message changed unexpectedly #### Description When defining discriminated unions with overlapping literal values, the error message format has changed. The error now shows a different message than expected when discriminator values map to multiple choices. #### Steps/Code to Reprodu...
pydantic__pydantic.acb0f10f.lm_rewrite__y6a8rjdp
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pydantic__pydantic.acb0f10f/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Discriminator value mapping error message changed unexpectedly #### Description When defining discrim...
49
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pydantic__pydantic.acb0f10f.lm_rewrite__y6a8rjdp.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/pydantic/_internal/_discriminated_union.py b/pydantic/_internal/_discriminated_union.py index 9daca32..fa468fa 100644 --- a/pydantic/_internal/_discriminated_union.py +++ b/pydantic/_internal/_discriminated_union.py @@ -469,7 +469,6 @@ class _ApplyInferredDiscriminator: existing_choice = self....
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_discriminated_union.py::test_distinct_choices" ], "failure": [] }, "PASS_TO_PASS": { "success": [ "te...
+ cd /testbed + : '>>>>> Start Test Output' + /root/.local/bin/uv run pytest --disable-warnings --color=no --tb=no --verbose Building pydantic @ file:///testbed Built pydantic @ file:///testbed Uninstalled 1 package in 1ms Installed 1 package in 7ms ============================= test session starts ===========...
756,473
2,941
8,644
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-oauthlib__oauthlib.1fd52536.func_basic__q4s72wrk
diff --git a/oauthlib/oauth1/rfc5849/endpoints/base.py b/oauthlib/oauth1/rfc5849/endpoints/base.py index 8d3d89c..8a3502e 100644 --- a/oauthlib/oauth1/rfc5849/endpoints/base.py +++ b/oauthlib/oauth1/rfc5849/endpoints/base.py @@ -28,26 +28,24 @@ class BaseEndpoint: """Extracts parameters from query, headers and...
[ "tests/oauth1/rfc5849/endpoints/test_access_token.py::AccessTokenEndpointTest::test_check_request_token", "tests/oauth1/rfc5849/endpoints/test_access_token.py::AccessTokenEndpointTest::test_check_verifier", "tests/oauth1/rfc5849/endpoints/test_access_token.py::AccessTokenEndpointTest::test_valid_request", "te...
[ "tests/oauth1/rfc5849/endpoints/test_base.py::BaseEndpointTest::test_missing_parameters", "tests/oauth1/rfc5849/endpoints/test_base.py::BaseEndpointTest::test_multiple_source_params", "tests/oauth1/rfc5849/endpoints/test_resource.py::ResourceEndpointTest::test_missing_parameters", "tests/oauth1/rfc5849/endpoi...
swebench/swesmith.x86_64.oauthlib_1776_oauthlib.1fd52536
swesmith/oauthlib__oauthlib.1fd52536
OAuth1 endpoint parameter extraction returns values in wrong order #### Description The `_get_signature_type_and_params` method in the OAuth1 base endpoint is returning values in the wrong order, causing signature validation and other OAuth operations to fail. #### Steps/Code to Reproduce ```python from oauthlib.oa...
oauthlib__oauthlib.1fd52536.func_basic__q4s72wrk
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/oauthlib__oauthlib.1fd52536/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> OAuth1 endpoint parameter extraction returns values in wrong order #### Description The `_get_signatu...
30
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/oauthlib__oauthlib.1fd52536.func_basic__q4s72wrk.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/oauthlib/oauth1/rfc5849/endpoints/base.py b/oauthlib/oauth1/rfc5849/endpoints/base.py index 8a3502e..d24d71d 100644 --- a/oauthlib/oauth1/rfc5849/endpoints/base.py +++ b/oauthlib/oauth1/rfc5849/endpoints/base.py @@ -62,7 +62,7 @@ class BaseEndpoint: raise errors.InvalidRequestError( ...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/oauth1/rfc5849/endpoints/test_access_token.py::AccessTokenEndpointTest::test_check_request_token", "tests/oauth1/rfc5849/endpoints/test_access_token.py::Acce...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
464,092
3,520
7,795
eog-gpt-5.5-2026-04-23-oh-task_20260121_183027_058_1afb323a_c53cb400
null
null
null
null
null
Could you please onboard a new support agent, Susan Adams (susan.adams075@servicenow.com)? She's joining the Customer Service team this week and will be taking over several open customer cases. Please create her user profile as active agent, and assign her location at Plot number 685, 'Ferguson Neck' Street, Mumbai, In...
task_20260121_183027_058_1afb323a_c53cb400
csm
# **CSM Assistant Policy** You are a Customer Service Management administrator. Your goal is to assist users in the Customer Service Management lifecycle by helping them register cases, validate entitlements, manage customer assets, raise escalations, attach relevant knowledge, close cases and in other related process...
Could you please onboard a new support agent, Susan Adams (susan.adams075@servicenow.com)? She's joining the Customer Service team this week and will be taking over several open customer cases. Please create her user profile as active agent, and assign her location at Plot number 685, 'Ferguson Neck' Street, Mumbai, In...
[ "find_locations", "find_user", "list_users", "assign_case_to_user", "get_cases_assigned_to", "update_user_details", "add_new_user", "add_new_group_member", "add_location" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-csm-server", "mcp_server_url": "http://localhost:8001", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/csm/dbs/db_1762254390925_u5icw4thh.sql", "context": {"x-user-email": "jose.roberson@servicenow.com"}, "user_info": {"user_id": 32, "name": "Jose Roberson", "email": "jose.roberson...
[{"verifier_type": "database_state", "name": "New Agent Created", "description": "Confirms that Susan Adams has been created and activated as an agent.", "gym_name": "sn-csm-server", "validation_config": {"query": "SELECT COUNT(*) FROM user WHERE first_name = 'Susan' AND last_name = 'Adams' AND email = 'susan.adams075@...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"csm","passed_verifiers":4,"resolved":false,"score":0,"total_verifiers":5,"verification_results":{"Cases Reassigned Correctly":{"actual":2,"comparison_type":"equals","details":"Comparison equals: 2 vs 2","expected":2,"passed":true,"query":"SELECT COUNT(*) FROM customer_case W...
null
Could you please onboard a new support agent, Susan Adams (susan.adams075@servicenow.com)? She's joining the Customer Service team this week and will be taking over several open customer cases. Please create her user profile as active agent, and assign her location at Plot number 685, 'Ferguson Neck' Street, Mumbai, In...
13
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-csm_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260121_183027_058_1afb323a_c53cb400.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
93,718
2,175
2,938
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-mozilla__bleach.73871d76.func_pm_remove_cond__kcyz9gdr
diff --git a/bleach/_vendor/html5lib/treewalkers/etree.py b/bleach/_vendor/html5lib/treewalkers/etree.py index a9d9450..3f4938a 100644 --- a/bleach/_vendor/html5lib/treewalkers/etree.py +++ b/bleach/_vendor/html5lib/treewalkers/etree.py @@ -101,11 +101,6 @@ def getETreeBuilder(ElementTreeImplementation): e...
[ "tests/test_clean.py::test_attributes_wildcard", "tests/test_clean.py::TestCleaner::test_basics", "tests/test_linkify.py::test_elements_inside_links", "tests/test_linkify.py::test_drop_link_tags", "tests/test_linkify.py::test_recognized_tags_arg" ]
[ "tests/test_callbacks.py::TestNofollowCallback::test_blank", "tests/test_callbacks.py::TestNofollowCallback::test_no_href", "tests/test_callbacks.py::TestNofollowCallback::test_basic", "tests/test_callbacks.py::TestNofollowCallback::test_mailto", "tests/test_callbacks.py::TestNofollowCallback::test_has_nofo...
swebench/swesmith.x86_64.mozilla_1776_bleach.73871d76
swesmith/mozilla__bleach.73871d76
# HTML tree walker breaks when traversing nested elements ## Description When using bleach to clean or linkify HTML content that contains nested elements, the tree walker fails to properly traverse the element tree. This causes issues with nested elements inside links and other complex HTML structures. ## Reproducti...
mozilla__bleach.73871d76.func_pm_remove_cond__kcyz9gdr
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/mozilla__bleach.73871d76/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> # HTML tree walker breaks when traversing nested elements ## Description When using bleach to clean or l...
45
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/mozilla__bleach.73871d76.func_pm_remove_cond__kcyz9gdr.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/bleach/_vendor/html5lib/treewalkers/etree.py b/bleach/_vendor/html5lib/treewalkers/etree.py index 3f4938a..a9d9450 100644 --- a/bleach/_vendor/html5lib/treewalkers/etree.py +++ b/bleach/_vendor/html5lib/treewalkers/etree.py @@ -101,6 +101,11 @@ def getETreeBuilder(ElementTreeImplementation): e...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_clean.py::test_attributes_wildcard", "tests/test_clean.py::TestCleaner::test_basics", "tests/test_linkify.py::test_elements_inside_links"...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,328,779
7,384
13,941
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20251221_222713_409_64a4ebb9_9915366d
null
null
null
null
null
Provision the new team 'Cloud_Ops_Alpha' managed by 'joanne.simpson@servicenow.com'. 1. Identify the manager in ITSM. 2. Create a new User Group named 'Cloud_Ops_Alpha' of type 'IT Support'. 3. Add the manager to this new group. 4. In Drive, create a shared folder named 'Cloud_Ops_Shared'. 5. Inside it, create a file ...
task_20251221_222713_409_64a4ebb9_9915366d
hybrid
You are an integrated automation agent for a hybrid environment managing both Google Drive and ServiceNow IT Service Management (ITSM). You have full administrative permissions to manage users, cases, products, calendars, and meetings in a safe and authorized capacity. Do not ask for confirmation before taking action.
Provision the new team 'Cloud_Ops_Alpha' managed by 'joanne.simpson@servicenow.com'. 1. Identify the manager in ITSM. 2. Create a new User Group named 'Cloud_Ops_Alpha' of type 'IT Support'. 3. Add the manager to this new group. 4. In Drive, create a shared folder named 'Cloud_Ops_Shared'. 5. Inside it, create a file ...
[ "get_user_using_email", "create_file", "update_user_group", "add_new_user_group", "add_new_group_member", "create_comments" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-google-drive-mcp", "mcp_server_url": "http://localhost:8009", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hybrid/dbs/db_1765211066177_klym6f8bt.sql", "context": {"x-auth-token": "ya29.E79oYmqohz4_yuNhAYfz7UsMx_YpFIXLCr1kMAgGtrvs9aINJEW2HfI54QtDpnvBx6yx7zaZkX9Y2Nekrf6VGmITgDUfwm...
[{"verifier_type": "database_state", "name": "Verify Group Created", "description": "Checks ITSM for the new User Group.", "gym_name": "gym-itsm-mcp", "validation_config": {"query": "SELECT COUNT(*) FROM user_group WHERE name = 'Cloud_Ops_Alpha' AND type = 'IT Support';", "expected_value": 1, "comparison_type": "equals...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hybrid","passed_verifiers":4,"resolved":false,"score":0,"total_verifiers":5,"verification_results":{"Verify Charter File":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(f.id) FROM files f JOIN fil...
null
Provision the new team 'Cloud_Ops_Alpha' managed by 'joanne.simpson@servicenow.com'. 1. Identify the manager in ITSM. 2. Create a new User Group named 'Cloud_Ops_Alpha' of type 'IT Support'. 3. Add the manager to this new group. 4. In Drive, create a shared folder named 'Cloud_Ops_Shared'. 5. Inside it, create a file ...
9
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hybrid_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251221_222713_409_64a4ebb9_9915366d.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
33,458
1,653
2,838
swes-gpt-5.5-2026-04-23-oh-pdfminer__pdfminer.six.1a8bd2f7.combine_file__db38faiu
diff --git a/pdfminer/layout.py b/pdfminer/layout.py index ee27b69..7f323e0 100644 --- a/pdfminer/layout.py +++ b/pdfminer/layout.py @@ -527,17 +527,17 @@ class LTTextLineHorizontal(LTTextLine): will be the same height as self, and also either left-, right-, or centrally-aligned. """ - ...
[ "tests/test_highlevel_extracttext.py::TestExtractText::test_issue_495_pdfobjref_iterable", "tests/test_highlevel_extracttext.py::TestExtractText::test_simple1_with_file", "tests/test_highlevel_extracttext.py::TestExtractText::test_simple1_with_string", "tests/test_highlevel_extracttext.py::TestExtractText::te...
[ "tests/test_converter.py::TestPaintPath::test_paint_path", "tests/test_converter.py::TestPaintPath::test_paint_path_mlllh", "tests/test_converter.py::TestPaintPath::test_paint_path_multiple_mlllh", "tests/test_converter.py::TestPaintPath::test_paint_path_quadrilaterals", "tests/test_converter.py::TestPaintP...
swebench/swesmith.x86_64.pdfminer_1776_pdfminer.six.1a8bd2f7
swesmith/pdfminer__pdfminer.six.1a8bd2f7
Text line neighbor finding returns incorrect results #### Description The `find_neighbors` method in `LTTextLineHorizontal` is not working correctly. When trying to find neighboring text lines that should be grouped together, it's returning unexpected results that don't match the expected alignment and height criteri...
pdfminer__pdfminer.six.1a8bd2f7.combine_file__db38faiu
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pdfminer__pdfminer.six.1a8bd2f7/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Text line neighbor finding returns incorrect results #### Description The `find_neighbors` method...
48
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pdfminer__pdfminer.six.1a8bd2f7.combine_file__db38faiu.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/pdfminer/layout.py b/pdfminer/layout.py index 7f323e0..ee5cfba 100644 --- a/pdfminer/layout.py +++ b/pdfminer/layout.py @@ -527,17 +527,17 @@ class LTTextLineHorizontal(LTTextLine): will be the same height as self, and also either left-, right-, or centrally-aligned. """ - ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_highlevel_extracttext.py::TestExtractText::test_simple3_with_file", "tests/test_highlevel_extracttext.py::TestExtractText::test_simple3_with_string", ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,038,740
2,220
12,792
swes-gpt-5.5-2026-04-23-oh-mozilla__bleach.73871d76.func_basic__p5orw6b9
diff --git a/bleach/html5lib_shim.py b/bleach/html5lib_shim.py index f083db7..22adbc2 100644 --- a/bleach/html5lib_shim.py +++ b/bleach/html5lib_shim.py @@ -310,13 +310,11 @@ class BleachHTMLTokenizer(HTMLTokenizer): def __init__(self, consume_entities=False, **kwargs): super().__init__(**kwargs) - ...
[ "tests/test_clean.py::test_character_entities_handling[&nbsp;-&nbsp;]", "tests/test_clean.py::test_character_entities_handling[http://example.com?active=true&current=true-http://example.com?active=true&amp;current=true]", "tests/test_clean.py::test_character_entities_handling[&#39;-&#39;]", "tests/test_clean....
[ "tests/test_callbacks.py::TestNofollowCallback::test_blank", "tests/test_callbacks.py::TestNofollowCallback::test_no_href", "tests/test_callbacks.py::TestNofollowCallback::test_basic", "tests/test_callbacks.py::TestNofollowCallback::test_mailto", "tests/test_callbacks.py::TestNofollowCallback::test_has_nofo...
swebench/swesmith.x86_64.mozilla_1776_bleach.73871d76
swesmith/mozilla__bleach.73871d76
HTML entity handling broken after recent changes #### Description HTML entities are not being processed correctly. When cleaning HTML content, entities like `&nbsp;`, `&amp;`, and numeric character references are getting double-escaped or not handled properly. #### Steps/Code to Reproduce ```python import bleach #...
mozilla__bleach.73871d76.func_basic__p5orw6b9
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/mozilla__bleach.73871d76/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> HTML entity handling broken after recent changes #### Description HTML entities are not being processed ...
43
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/mozilla__bleach.73871d76.func_basic__p5orw6b9.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/bleach/html5lib_shim.py b/bleach/html5lib_shim.py index 22adbc2..0ed1180 100644 --- a/bleach/html5lib_shim.py +++ b/bleach/html5lib_shim.py @@ -310,7 +310,7 @@ class BleachHTMLTokenizer(HTMLTokenizer): def __init__(self, consume_entities=False, **kwargs): super().__init__(**kwargs) - ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_clean.py::test_character_entities_handling[&nbsp;-&nbsp;]", "tests/test_clean.py::test_character_entities_handling[http://example.com?active=true&curren...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,987,503
3,322
13,261
eog-gpt-5.5-2026-04-23-oh-task_20251225_071630_879_701c5774_99b636e9
null
null
null
null
null
I have an active work email address: jessicabrown@servicenow.com. I need a send-as alias configured for this address, with replies directed to my personal primary email. The display name should match my first and last name as listed in the HR system. The email signature should read: "Thank you! Your agent {First name} ...
task_20251225_071630_879_701c5774_99b636e9
hybrid
# **Email Management Assistant Policy** **Role:** Email Management Assistant **Mandate:** Secure, accurate, and efficient management of user email communications (Messages, Drafts, Threads, Labels, Filters, and Settings). You must act solely based on **confirmed user context** and **existing data integrity rules**. ...
I have an active work email address: jessicabrown@servicenow.com. I need a send-as alias configured for this address, with replies directed to my personal primary email. The display name should match my first and last name as listed in the HR system. The email signature should read: "Thank you! Your agent {First name} ...
[ "list_users", "send_message", "create_send_as_alias", "delete_send_as_alias", "verify_send_as_alias", "get_user_using_email", "list_send_as_aliases" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hybrid/dbs/db_1765361971944_6ld5uejbx.sql", "context": {"x-email-user-token": "Kx5XwZK6imPxgyAE88USnS4lF3np3br81Tf9FZr4Vno", "user_id": "104daf4b-2023-49f1-8def-39b4e9dcda5e", "pr...
[{"verifier_type": "database_state", "name": "Check that the message is sent as requested", "description": "Check that the message is sent as requested", "gym_name": "gym-email-mcp", "validation_config": {"query": "select count(*) from messages m join message_parts mp on mp.message_id = m.id join message_headers mh on ...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hybrid","passed_verifiers":4,"resolved":false,"score":0,"total_verifiers":8,"verification_results":{"Check that the message is sent as requested":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 3","expected":3,"passed":false,"query":"select count(*)...
null
I have an active work email address: jessicabrown@servicenow.com. I need a send-as alias configured for this address, with replies directed to my personal primary email. The display name should match my first and last name as listed in the HR system. The email signature should read: "Thank you! Your agent {First name} ...
11
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hybrid_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251225_071630_879_701c5774_99b636e9.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
117,537
7,930
8,969
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-mozilla__bleach.73871d76.func_pm_class_rm_funcs__csgky3oo
diff --git a/bleach/_vendor/html5lib/html5parser.py b/bleach/_vendor/html5lib/html5parser.py index 5427b7d..37dfc4a 100644 --- a/bleach/_vendor/html5lib/html5parser.py +++ b/bleach/_vendor/html5lib/html5parser.py @@ -1010,9 +1010,6 @@ def getPhases(debug): self.tree.reconstructActiveFormattingElements() ...
[ "tests/test_clean.py::test_clean_idempotent[text<em>]", "tests/test_clean.py::test_clean_idempotent_img", "tests/test_clean.py::test_only_text_is_cleaned", "tests/test_clean.py::test_empty", "tests/test_clean.py::test_content_has_no_html", "tests/test_clean.py::test_html_is_lowercased", "tests/test_clea...
[ "tests/test_callbacks.py::TestNofollowCallback::test_blank", "tests/test_callbacks.py::TestNofollowCallback::test_no_href", "tests/test_callbacks.py::TestNofollowCallback::test_basic", "tests/test_callbacks.py::TestNofollowCallback::test_mailto", "tests/test_callbacks.py::TestNofollowCallback::test_has_nofo...
swebench/swesmith.x86_64.mozilla_1776_bleach.73871d76
swesmith/mozilla__bleach.73871d76
HTML parser methods missing after recent changes #### Description After some recent changes to the HTML parser, several methods that were previously available are now missing, causing `NameError` exceptions when trying to process HTML content. #### Steps/Code to Reproduce ```python import bleach # This should work...
mozilla__bleach.73871d76.func_pm_class_rm_funcs__csgky3oo
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/mozilla__bleach.73871d76/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> HTML parser methods missing after recent changes #### Description After some recent changes to the HTML ...
59
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/mozilla__bleach.73871d76.func_pm_class_rm_funcs__csgky3oo.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/bleach/_vendor/html5lib/html5parser.py b/bleach/_vendor/html5lib/html5parser.py index 37dfc4a..5427b7d 100644 --- a/bleach/_vendor/html5lib/html5parser.py +++ b/bleach/_vendor/html5lib/html5parser.py @@ -1010,6 +1010,9 @@ def getPhases(debug): self.tree.reconstructActiveFormattingElements() ...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_clean.py::test_clean_idempotent[text<em>]", "tests/test_clean.py::test_clean_idempotent_img", "tests/test_clean.py::test_only_text_is_cle...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,754,629
10,074
20,534
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-bottlepy__bottle.a8dfef30.func_pm_class_rm_funcs__weo8gf3c
diff --git a/bottle.py b/bottle.py index 286b5bc..52ef6c4 100755 --- a/bottle.py +++ b/bottle.py @@ -1146,13 +1146,6 @@ class BaseRequest(object): #: Maximum size of memory buffer for :attr:`body` in bytes. MEMFILE_MAX = 102400 - def __init__(self, environ=None): - """ Wrap a WSGI environ dictiona...
[ "test/test_auth.py::TestBasicAuth::test__header", "test/test_contextlocals.py::TestThreadLocals::test_request", "test/test_environ.py::TestRequest::test_app_property", "test/test_environ.py::TestRequest::test_auth", "test/test_environ.py::TestRequest::test_bigbody", "test/test_environ.py::TestRequest::tes...
[ "test/test_app.py::TestApplicationObject::test_setattr", "test/test_config.py::TestConfDict::test_gc_overlays", "test/test_config.py::TestConfDict::test_isadict", "test/test_config.py::TestConfDict::test_load_dict", "test/test_config.py::TestConfDict::test_load_module", "test/test_config.py::TestConfDict:...
swebench/swesmith.x86_64.bottlepy_1776_bottle.a8dfef30
swesmith/bottlepy__bottle.a8dfef30
BaseRequest missing __init__ method causes TypeError on instantiation #### Description After recent changes, creating a BaseRequest instance fails with a TypeError because the `__init__` method has been removed. #### Steps/Code to Reproduce ```python import bottle # This should work but now raises TypeError reques...
bottlepy__bottle.a8dfef30.func_pm_class_rm_funcs__weo8gf3c
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/bottlepy__bottle.a8dfef30/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> BaseRequest missing __init__ method causes TypeError on instantiation #### Description After recent cha...
46
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/bottlepy__bottle.a8dfef30.func_pm_class_rm_funcs__weo8gf3c.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/bottle.py b/bottle.py index 52ef6c4..41f28c0 100755 --- a/bottle.py +++ b/bottle.py @@ -1146,6 +1146,13 @@ class BaseRequest(object): #: Maximum size of memory buffer for :attr:`body` in bytes. MEMFILE_MAX = 102400 + def __init__(self, environ=None): + """ Wrap a WSGI environ dictiona...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/test_contextlocals.py::TestThreadLocals::test_request", "test/test_environ.py::TestRequest::test_app_property", "test/test_environ.py::TestReq...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,018,814
4,060
8,908
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20251203_194320_796_df2c536d_c885488e
null
null
null
null
null
Wipe my pry calendar and update it to have the same color as my personal one. Add a confirmed 2026 Strategy Kickoff event for December 1st, 2025, from 9:00 AM, to last for 2 hrs with a description, 'Mandatory attendance for all leads.' Attach the 2026 strategy planning document (https://docs.google.com/document/d/techn...
task_20251203_194320_796_df2c536d_c885488e
calendar
You are a Google Calendar automation agent with full administrative permissions to manage users, meetings, recordings, and messages. Operate in a safe and fully authorized environment — you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform...
Wipe my pry calendar and update it to have the same color as my personal one. Add a confirmed 2026 Strategy Kickoff event for December 1st, 2025, from 9:00 AM, to last for 2 hrs with a description, 'Mandatory attendance for all leads.' Attach the 2026 strategy planning document (https://docs.google.com/document/d/techn...
[ "watch_events", "get_calendar_list", "clear_calendar", "update_calendar_in_list", "create_event" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1764361046464_47vxzhsli.sql", "context": {"x-access-token": "ya29.A0ARrdaM-Zx8Nw3Q4pVb6Ls9R1mT0cG2uF5yH7kJd8sA1Lq2.wErtYuIoPaSdFgHjKlZxCvBnM987_65-43"}, "user_info"...
[{"verifier_type": "database_state", "name": "strategy_kickoff_created", "description": "strategy_kickoff_created", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) FROM events e WHERE e.summary = '2026 Strategy Kickoff' AND e.status = 'confirmed' AND e.description = 'Mandatory attendance for...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":3,"resolved":false,"score":0,"total_verifiers":4,"verification_results":{"check_attendees":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 3","expected":3,"passed":false,"query":"SELECT COUNT(*) FROM attendees WHERE even...
null
Wipe my pry calendar and update it to have the same color as my personal one. Add a confirmed 2026 Strategy Kickoff event for December 1st, 2025, from 9:00 AM, to last for 2 hrs with a description, 'Mandatory attendance for all leads.' Attach the 2026 strategy planning document (https://docs.google.com/document/d/techn...
9
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251203_194320_796_df2c536d_c885488e.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
58,577
2,033
3,982
swes-gemini-gemini-3.5-flash-oh-paramiko__paramiko.23f92003.func_basic__vtaqnh1i
diff --git a/paramiko/pipe.py b/paramiko/pipe.py index 65944fa..9bfcf46 100644 --- a/paramiko/pipe.py +++ b/paramiko/pipe.py @@ -143,6 +143,6 @@ def make_or_pipe(pipe): """ p1 = OrPipe(pipe) p2 = OrPipe(pipe) - p1._partner = p2 - p2._partner = p1 - return p1, p2 + p1._partner = p1 + p2._pa...
[ "tests/test_buffered_pipe.py::BufferedPipeTest::test_or_pipe", "tests/test_transport.py::TransportTest::test_select", "tests/test_transport.py::TransportTest::test_select_after_close", "tests/test_transport.py::TransportTest::test_stderr_select", "tests/test_transport.py::ServiceRequestingTransportTest::tes...
[ "tests/agent.py::AgentKey_::str_is_repr", "tests/agent.py::AgentKey_::init::needs_at_least_two_arguments", "tests/agent.py::AgentKey_::init::sets_attributes_and_parses_blob", "tests/agent.py::AgentKey_::init::comment_optional", "tests/agent.py::AgentKey_::init::sets_inner_key_when_known_type[rsa]", "tests...
swebench/swesmith.x86_64.paramiko_1776_paramiko.23f92003
swesmith/paramiko__paramiko.23f92003
OrPipe partners incorrectly assigned causing select() failures When using `make_or_pipe()`, the partner assignment is incorrect which breaks the pipe functionality. This causes issues with channel selection operations. #### Steps/Code to Reproduce ```python import paramiko.pipe as pipe # Create an or pipe p = pipe....
paramiko__paramiko.23f92003.func_basic__vtaqnh1i
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/paramiko__paramiko.23f92003/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> OrPipe partners incorrectly assigned causing select() failures When using `make_or_pipe()`, the partne...
40
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/paramiko__paramiko.23f92003.func_basic__vtaqnh1i.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/paramiko/pipe.py b/paramiko/pipe.py index 9bfcf46..65944fa 100644 --- a/paramiko/pipe.py +++ b/paramiko/pipe.py @@ -143,6 +143,6 @@ def make_or_pipe(pipe): """ p1 = OrPipe(pipe) p2 = OrPipe(pipe) - p1._partner = p1 - p2._partner = p2 - return p2, p1 + p1._partner = p2 + p2._pa...
gemini/gemini-3.5-flash
{ "timed_out": true, "timeout": 90 }
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
920,265
6,818
15,640
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-mozilla__bleach.73871d76.combine_file__5vejqplu
diff --git a/bleach/_vendor/html5lib/_tokenizer.py b/bleach/_vendor/html5lib/_tokenizer.py index d884801..be3c6e9 100644 --- a/bleach/_vendor/html5lib/_tokenizer.py +++ b/bleach/_vendor/html5lib/_tokenizer.py @@ -856,16 +856,16 @@ class HTMLTokenizer(object): def beforeAttributeNameState(self): data = s...
[ "tests/test_clean.py::test_html_is_lowercased", "tests/test_clean.py::test_comments[<!--comment-->text-True-text]", "tests/test_clean.py::test_comments[<!--comment-->text-False-<!--comment-->text]", "tests/test_clean.py::test_comments[text<!--comment-->-False-text<!--comment-->]", "tests/test_clean.py::test...
[ "tests/test_callbacks.py::TestNofollowCallback::test_blank", "tests/test_callbacks.py::TestNofollowCallback::test_no_href", "tests/test_callbacks.py::TestNofollowCallback::test_basic", "tests/test_callbacks.py::TestNofollowCallback::test_mailto", "tests/test_callbacks.py::TestNofollowCallback::test_has_nofo...
swebench/swesmith.x86_64.mozilla_1776_bleach.73871d76
swesmith/mozilla__bleach.73871d76
HTML tokenizer produces empty output for many inputs #### Description The HTML tokenizer is returning empty strings for various HTML inputs that should produce valid output. This affects both cleaning and linkifying operations. #### Steps/Code to Reproduce ```python import bleach # Basic cleaning should preserve c...
mozilla__bleach.73871d76.combine_file__5vejqplu
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/mozilla__bleach.73871d76/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> HTML tokenizer produces empty output for many inputs #### Description The HTML tokenizer is returning em...
83
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/mozilla__bleach.73871d76.combine_file__5vejqplu.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/bleach/_vendor/html5lib/_tokenizer.py b/bleach/_vendor/html5lib/_tokenizer.py index be3c6e9..d884801 100644 --- a/bleach/_vendor/html5lib/_tokenizer.py +++ b/bleach/_vendor/html5lib/_tokenizer.py @@ -856,16 +856,16 @@ class HTMLTokenizer(object): def beforeAttributeNameState(self): data = s...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_clean.py::test_html_is_lowercased", "tests/test_clean.py::test_comments[<!--comment-->text-True-text]", "tests/test_clean.py::test_commen...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
3,197,888
5,275
17,947
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20260109_063639_321_8e9e30d7_c31290aa
null
null
null
null
null
I need to reorganize our TechCorp Solutions Team so create a private Leadership channel which will becomes the official repository for all formal leadership decisions, meeting minutes, and policy documentation. Right now, things are scattered - people upload critical files in message threads, and there's confusion abou...
task_20260109_063639_321_8e9e30d7_c31290aa
teams
# **Teams Assistant Policy** You are a **Microsoft Teams Management Assistant**. Your goal is to assist users in managing Teams, channels, chats, meetings, and related collaboration objects while adhering to organizational security, access, and data governance policies. You must always act on **confirmed context** a...
I need to reorganize our TechCorp Solutions Team so create a private Leadership channel which will becomes the official repository for all formal leadership decisions, meeting minutes, and policy documentation. Right now, things are scattered - people upload critical files in message threads, and there's confusion abou...
[ "list_teams_apps", "list_teams", "list_users", "update_channel_tab", "add_channel_member", "send_channel_message", "list_channels", "create_teamwork_tag", "list_tabs_in_channel", "create_channel", "add_tabs_to_channels" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-teams-mcp", "mcp_server_url": "http://localhost:8002", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/teams/dbs/db_1764136986337_rjetcjsa2.sql", "context": {"x-teams-access-token": "ya29.A0ARrdaM-super999unique888token777special666.SuperUserAllPermissionsTokenAccess_99-99"}, "user...
[{"verifier_type": "database_state", "name": "planner_tab_added_to_leadership", "description": "Confirm Planner tab (Leadership Action Items) was added to Leadership channel.", "gym_name": "gym-teams-mcp", "validation_config": {"query": "SELECT COUNT(*) FROM teams_channel_tabs tct JOIN channels c ON c.channel_id = tct....
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"teams","passed_verifiers":11,"resolved":false,"score":0,"total_verifiers":12,"verification_results":{"announcement_messages_with_sponsors":{"actual":2,"comparison_type":"equals","details":"Comparison equals: 2 vs 2","expected":2,"passed":true,"query":"SELECT COUNT(*) FROM ch...
null
I need to reorganize our TechCorp Solutions Team so create a private Leadership channel which will becomes the official repository for all formal leadership decisions, meeting minutes, and policy documentation. Right now, things are scattered - people upload critical files in message threads, and there's confusion abou...
20
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-teams_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260109_063639_321_8e9e30d7_c31290aa.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
144,520
3,271
8,037
swes-gpt-5.5-2026-04-23-oh-pydantic__pydantic.acb0f10f.pr_9561
diff --git a/pydantic/alias_generators.py b/pydantic/alias_generators.py index 0b7653f..dbd0637 100644 --- a/pydantic/alias_generators.py +++ b/pydantic/alias_generators.py @@ -4,10 +4,6 @@ import re __all__ = ('to_pascal', 'to_camel', 'to_snake') -# TODO: in V3, change the argument names to be more descriptive -#...
[ "tests/test_utils.py::test_to_camel_from_camel" ]
[ "tests/benchmarks/test_attribute_access.py::test_setattr", "tests/benchmarks/test_attribute_access.py::test_getattr", "tests/benchmarks/test_discriminated_unions.py::test_schema_build", "tests/benchmarks/test_discriminated_unions.py::test_efficiency_with_highly_nested_examples", "tests/benchmarks/test_fasta...
swebench/swesmith.x86_64.pydantic_1776_pydantic.acb0f10f
swesmith/pydantic__pydantic.acb0f10f
# Bug: to_camel() function doesn't preserve existing camelCase strings ## Description The `to_camel()` function in the alias_generators module is not preserving strings that are already in camelCase format. Instead, it's converting them to a different format. ## Code for reproduction ```python from pydantic.alias_g...
pydantic__pydantic.acb0f10f.pr_9561
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pydantic__pydantic.acb0f10f/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> # Bug: to_camel() function doesn't preserve existing camelCase strings ## Description The `to_camel()...
34
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pydantic__pydantic.acb0f10f.pr_9561.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/pydantic/alias_generators.py b/pydantic/alias_generators.py index dbd0637..48d5b13 100644 --- a/pydantic/alias_generators.py +++ b/pydantic/alias_generators.py @@ -27,6 +27,9 @@ def to_camel(snake: str) -> str: Returns: The converted camelCase string. """ + if re.match('^[a-z]+[A-Za-z...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_utils.py::test_to_camel_from_camel" ], "failure": [] }, "PASS_TO_PASS": { "success": [ "tests/benchma...
+ cd /testbed + : '>>>>> Start Test Output' + /root/.local/bin/uv run pytest --disable-warnings --color=no --tb=no --verbose Building pydantic @ file:///testbed Built pydantic @ file:///testbed Uninstalled 1 package in 1ms Installed 1 package in 7ms ============================= test session starts ===========...
543,655
3,224
8,669
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-dask__dask.5f61e423.lm_rewrite__xza8r48g
diff --git a/dask/dataframe/io/parquet/arrow.py b/dask/dataframe/io/parquet/arrow.py index c4e02d6..6a89f01 100644 --- a/dask/dataframe/io/parquet/arrow.py +++ b/dask/dataframe/io/parquet/arrow.py @@ -766,37 +766,46 @@ class ArrowDatasetEngine(Engine): return i_offset, full_metadata, metadata_file_exists, extr...
[ "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_parquet_len[arrow]", "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_parquet_missing_stats[arrow]", "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_parquet_all_na_column[arrow-nan]", "dask/dataframe/dask_expr/io/tests/test_parquet.py::...
[ "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_parquet_len_filter[arrow]", "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_aggregate_statistics_to_file", "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_combine_statistics", "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_in...
swebench/swesmith.x86_64.dask_1776_dask.5f61e423
swesmith/dask__dask.5f61e423
### KeyError: 'distributed' when reading Parquet files #### Description It seems there's an issue when attempting to read Parquet files using the `pyarrow` engine. The error `KeyError: 'distributed'` is raised during the read operation. This occurs in various scenarios, including when using filters or reading partiti...
dask__dask.5f61e423.lm_rewrite__xza8r48g
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/dask__dask.5f61e423/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> ### KeyError: 'distributed' when reading Parquet files #### Description It seems there's an issue when attemp...
100
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/dask__dask.5f61e423.lm_rewrite__xza8r48g.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [], "failure": [ "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_parquet_len[arrow]", "dask/dataframe/dask_expr/io/tests/test_parquet.py::test_pa...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
3,558,160
14,581
26,922
swes-gemini-gemini-3.5-flash-oh-pdfminer__pdfminer.six.1a8bd2f7.combine_file__db38faiu
diff --git a/pdfminer/layout.py b/pdfminer/layout.py index ee27b69..7f323e0 100644 --- a/pdfminer/layout.py +++ b/pdfminer/layout.py @@ -527,17 +527,17 @@ class LTTextLineHorizontal(LTTextLine): will be the same height as self, and also either left-, right-, or centrally-aligned. """ - ...
[ "tests/test_highlevel_extracttext.py::TestExtractText::test_issue_495_pdfobjref_iterable", "tests/test_highlevel_extracttext.py::TestExtractText::test_simple1_with_file", "tests/test_highlevel_extracttext.py::TestExtractText::test_simple1_with_string", "tests/test_highlevel_extracttext.py::TestExtractText::te...
[ "tests/test_converter.py::TestPaintPath::test_paint_path", "tests/test_converter.py::TestPaintPath::test_paint_path_mlllh", "tests/test_converter.py::TestPaintPath::test_paint_path_multiple_mlllh", "tests/test_converter.py::TestPaintPath::test_paint_path_quadrilaterals", "tests/test_converter.py::TestPaintP...
swebench/swesmith.x86_64.pdfminer_1776_pdfminer.six.1a8bd2f7
swesmith/pdfminer__pdfminer.six.1a8bd2f7
Text line neighbor finding returns incorrect results #### Description The `find_neighbors` method in `LTTextLineHorizontal` is not working correctly. When trying to find neighboring text lines that should be grouped together, it's returning unexpected results that don't match the expected alignment and height criteri...
pdfminer__pdfminer.six.1a8bd2f7.combine_file__db38faiu
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pdfminer__pdfminer.six.1a8bd2f7/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Text line neighbor finding returns incorrect results #### Description The `find_neighbors` method...
43
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pdfminer__pdfminer.six.1a8bd2f7.combine_file__db38faiu.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/pdfminer/layout.py b/pdfminer/layout.py index 7f323e0..5acaece 100644 --- a/pdfminer/layout.py +++ b/pdfminer/layout.py @@ -527,17 +527,17 @@ class LTTextLineHorizontal(LTTextLine): will be the same height as self, and also either left-, right-, or centrally-aligned. """ - ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_highlevel_extracttext.py::TestExtractText::test_simple4_with_file", "tests/test_highlevel_extracttext.py::TestExtractText::test_simple4_with_string", ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,133,001
9,647
19,023
swes-gemini-gemini-3.5-flash-oh-amueller__word_cloud.ec24191c.func_basic__tcz41609
diff --git a/wordcloud/wordcloud.py b/wordcloud/wordcloud.py index 1622c03..1c598ee 100644 --- a/wordcloud/wordcloud.py +++ b/wordcloud/wordcloud.py @@ -650,10 +650,10 @@ class WordCloud(object): def to_image(self): self._check_generated() if self.mask is not None: - width = self.mask....
[ "test/test_wordcloud.py::test_default", "test/test_wordcloud.py::test_writing_to_file", "test/test_wordcloud.py::test_recolor", "test/test_wordcloud.py::test_mask", "test/test_wordcloud.py::test_mask_contour", "test/test_wordcloud_cli.py::test_cli_writes_to_imagefile", "test/test_wordcloud_cli.py::test_...
[ "test/test_wordcloud.py::test_collocations", "test/test_wordcloud.py::test_collocation_stopwords", "test/test_wordcloud.py::test_collocation_stopwords_uppercase", "test/test_wordcloud.py::test_plurals_numbers", "test/test_wordcloud.py::test_multiple_s", "test/test_wordcloud.py::test_empty_text", "test/t...
swebench/swesmith.x86_64.amueller_1776_word_cloud.ec24191c
swesmith/amueller__word_cloud.ec24191c
WordCloud.to_image() returns None instead of PIL Image #### Description The `to_image()` method is returning `None` instead of a PIL Image object, breaking existing functionality that depends on the returned image. #### Steps/Code to Reproduce ```python from wordcloud import WordCloud # Create a simple word cloud ...
amueller__word_cloud.ec24191c.func_basic__tcz41609
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/amueller__word_cloud.ec24191c/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> WordCloud.to_image() returns None instead of PIL Image #### Description The `to_image()` method is ...
45
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/amueller__word_cloud.ec24191c.func_basic__tcz41609.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/wordcloud/wordcloud.py b/wordcloud/wordcloud.py index 1c598ee..5dc7317 100644 --- a/wordcloud/wordcloud.py +++ b/wordcloud/wordcloud.py @@ -668,7 +668,7 @@ class WordCloud(object): int(position[1] * self.scale)) draw.text(pos, word, fill=color, font=transposed_font) - ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/test_wordcloud.py::test_default", "test/test_wordcloud.py::test_writing_to_file", "test/test_wordcloud.py::test_recolor", "tes...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
963,708
7,841
16,242
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-stanfordnlp__string2string.c4a72f59.func_pm_op_change_const__8gw8ztxj
diff --git a/string2string/alignment/classical.py b/string2string/alignment/classical.py index 86c8039..416f414 100644 --- a/string2string/alignment/classical.py +++ b/string2string/alignment/classical.py @@ -661,16 +661,16 @@ class Hirschberg(NeedlemanWunsch): # Initialize the first row of the score matrix. ...
[ "tests/test_alignment.py::AlignmentTestCase::test_hiirschberg" ]
[ "tests/test_alignment.py::AlignmentTestCase::test_dtw", "tests/test_alignment.py::AlignmentTestCase::test_longest_common_subsequence", "tests/test_alignment.py::AlignmentTestCase::test_longest_common_subsubtring", "tests/test_alignment.py::AlignmentTestCase::test_needleman_wunsch", "tests/test_alignment.py:...
swebench/swesmith.x86_64.stanfordnlp_1776_string2string.c4a72f59
swesmith/stanfordnlp__string2string.c4a72f59
### Description It seems there's an issue with the Hirschberg alignment algorithm after the recent changes. When attempting to align certain sequences, the algorithm throws an unexpected error. This behavior is inconsistent with the expected output, where the sequences should align correctly without any errors. ### S...
stanfordnlp__string2string.c4a72f59.func_pm_op_change_const__8gw8ztxj
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/stanfordnlp__string2string.c4a72f59/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> ### Description It seems there's an issue with the Hirschberg alignment algorithm after the re...
51
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/stanfordnlp__string2string.c4a72f59.func_pm_op_change_const__8gw8ztxj.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/string2string/alignment/classical.py b/string2string/alignment/classical.py index 416f414..cb8cd62 100644 --- a/string2string/alignment/classical.py +++ b/string2string/alignment/classical.py @@ -663,14 +663,14 @@ class Hirschberg(NeedlemanWunsch): score_matrix[0, j] = score_matrix[0, j - 1] +...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_alignment.py::AlignmentTestCase::test_hiirschberg" ], "failure": [] }, "PASS_TO_PASS": { "success": [ ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,571,843
10,647
20,602
swes-gemini-gemini-3.5-flash-oh-andialbrecht__sqlparse.e57923b3.func_pm_ctrl_shuffle__6vfrll59
diff --git a/sqlparse/filters/others.py b/sqlparse/filters/others.py index 6c1680e..8febd26 100644 --- a/sqlparse/filters/others.py +++ b/sqlparse/filters/others.py @@ -121,12 +121,11 @@ class StripWhitespaceFilter: self._stripws_default(tlist) def process(self, stmt, depth=0): - [self.process(sg...
[ "tests/test_format.py::TestFormat::test_strip_comments_multi", "tests/test_format.py::TestFormat::test_strip_ws", "tests/test_format.py::TestFormatReindentAligned::test_basic", "tests/test_format.py::TestFormatReindentAligned::test_case_statement", "tests/test_format.py::TestFormatReindentAligned::test_case...
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_valid_args", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::t...
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
StripWhitespaceFilter not processing sublists after whitespace removal Description When using the StripWhitespaceFilter, sublists within SQL statements are not being processed correctly. The filter appears to return early after stripping whitespace from the main statement, preventing proper processing of nested struc...
andialbrecht__sqlparse.e57923b3.func_pm_ctrl_shuffle__6vfrll59
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> StripWhitespaceFilter not processing sublists after whitespace removal Description When using the...
40
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_pm_ctrl_shuffle__6vfrll59.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/AGENTS.md b/AGENTS.md new file mode 100644 index 0000000..1bc0ee9 --- /dev/null +++ b/AGENTS.md @@ -0,0 +1,12 @@ +# sqlparse Repository Skill & Context + +## Structure & Architecture +- `sqlparse` parses SQL statements into a tree of tokens represented by classes in `sqlparse/sql.py`. +- Elements that cont...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_format.py::TestFormat::test_strip_comments_multi", "tests/test_format.py::TestFormat::test_strip_ws", "tests/test_format.py::TestFormatRe...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
913,461
9,642
17,990
swes-gpt-5.5-2026-04-23-oh-amueller__word_cloud.ec24191c.func_basic__tcz41609
diff --git a/wordcloud/wordcloud.py b/wordcloud/wordcloud.py index 1622c03..1c598ee 100644 --- a/wordcloud/wordcloud.py +++ b/wordcloud/wordcloud.py @@ -650,10 +650,10 @@ class WordCloud(object): def to_image(self): self._check_generated() if self.mask is not None: - width = self.mask....
[ "test/test_wordcloud.py::test_default", "test/test_wordcloud.py::test_writing_to_file", "test/test_wordcloud.py::test_recolor", "test/test_wordcloud.py::test_mask", "test/test_wordcloud.py::test_mask_contour", "test/test_wordcloud_cli.py::test_cli_writes_to_imagefile", "test/test_wordcloud_cli.py::test_...
[ "test/test_wordcloud.py::test_collocations", "test/test_wordcloud.py::test_collocation_stopwords", "test/test_wordcloud.py::test_collocation_stopwords_uppercase", "test/test_wordcloud.py::test_plurals_numbers", "test/test_wordcloud.py::test_multiple_s", "test/test_wordcloud.py::test_empty_text", "test/t...
swebench/swesmith.x86_64.amueller_1776_word_cloud.ec24191c
swesmith/amueller__word_cloud.ec24191c
WordCloud.to_image() returns None instead of PIL Image #### Description The `to_image()` method is returning `None` instead of a PIL Image object, breaking existing functionality that depends on the returned image. #### Steps/Code to Reproduce ```python from wordcloud import WordCloud # Create a simple word cloud ...
amueller__word_cloud.ec24191c.func_basic__tcz41609
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/amueller__word_cloud.ec24191c/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> WordCloud.to_image() returns None instead of PIL Image #### Description The `to_image()` method is ...
31
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/amueller__word_cloud.ec24191c.func_basic__tcz41609.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/wordcloud/wordcloud.py b/wordcloud/wordcloud.py index 1c598ee..5dc7317 100644 --- a/wordcloud/wordcloud.py +++ b/wordcloud/wordcloud.py @@ -668,7 +668,7 @@ class WordCloud(object): int(position[1] * self.scale)) draw.text(pos, word, fill=color, font=transposed_font) - ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/test_wordcloud.py::test_default", "test/test_wordcloud.py::test_writing_to_file", "test/test_wordcloud.py::test_recolor", "tes...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
483,223
1,374
7,988
eog-gemini-gemini-3.5-flash-oh-task_20260105_074654_229_bca89e7d_39d9c8ac
null
null
null
null
null
Create a calendar named 'Hackathon 2026' in the UTC time, ensuring it uses a color ID clearly different from all existing calendars for easy identification. Schedule a 48-hour all-day event titled 'Build, Ship, Win!' from March 20 to March 22. Also, schedule a Google Meet 'Hackathon Kickoff Session' at 09:00 UTC on Mar...
task_20260105_074654_229_bca89e7d_39d9c8ac
calendar
You are a Google Calendar automation agent with full administrative permissions to manage users, meetings, recordings, and messages. Operate in a safe and fully authorized environment — you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform...
Create a calendar named 'Hackathon 2026' in the UTC time, ensuring it uses a color ID clearly different from all existing calendars for easy identification. Schedule a 48-hour all-day event titled 'Build, Ship, Win!' from March 20 to March 22. Also, schedule a Google Meet 'Hackathon Kickoff Session' at 09:00 UTC on Mar...
[ "list_events", "create_calendar", "add_calendar_to_list", "create_event", "create_event", "get_colors", "get_calendar_list", "watch_events", "create_event" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1762868439331_kf914hbmw.sql", "context": {"x-access-token": "ya29.A0ARrdaM-k9Vq7GzY2pL4mQf8sN1xT0bR3uHcJWv5yKzP6eF2.qwErTyUIopASDfGhJkLzXcVbNm12_34-56"}, "user_info...
[{"verifier_type": "database_state", "name": "Calendar Created", "description": "Check calendar creation with unique color", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) AS count FROM calendars WHERE user_id = 'alice_manager' AND summary = 'Hackathon 2026' AND time_zone = 'UTC' AND color_...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":6,"resolved":true,"score":1,"total_verifiers":6,"verification_results":{"Alert Setup":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) AS count FROM watch_channels WH...
null
Create a calendar named 'Hackathon 2026' in the UTC time, ensuring it uses a color ID clearly different from all existing calendars for easy identification. Schedule a 48-hour all-day event titled 'Build, Ship, Win!' from March 20 to March 22. Also, schedule a Google Meet 'Hackathon Kickoff Session' at 09:00 UTC on Mar...
14
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260105_074654_229_bca89e7d_39d9c8ac.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gemini/gemini-3.5-flash
null
null
216,640
6,429
7,898
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-bottlepy__bottle.a8dfef30.func_pm_remove_cond__wtiszsqd
diff --git a/bottle.py b/bottle.py index 286b5bc..8f6a686 100755 --- a/bottle.py +++ b/bottle.py @@ -1694,18 +1694,10 @@ class BaseResponse(object): def _set_status(self, status): if isinstance(status, int): code, status = status, _HTTP_STATUS_LINES.get(status) - elif ' ' in status: - ...
[ "test/test_environ.py::TestResponse::test_clone_cookie", "test/test_environ.py::TestResponse::test_constructor_status", "test/test_environ.py::TestResponse::test_set_status", "test/test_environ.py::TestRedirect::test_absolute_path", "test/test_environ.py::TestRedirect::test_host_http_1_0", "test/test_envi...
[ "test/test_app.py::TestApplicationObject::test_setattr", "test/test_auth.py::TestBasicAuth::test__header", "test/test_config.py::TestConfDict::test_gc_overlays", "test/test_config.py::TestConfDict::test_isadict", "test/test_config.py::TestConfDict::test_load_dict", "test/test_config.py::TestConfDict::test...
swebench/swesmith.x86_64.bottlepy_1776_bottle.a8dfef30
swesmith/bottlepy__bottle.a8dfef30
Status line validation removed causing UnboundLocalError When setting response status, the validation logic for status strings was removed, causing an UnboundLocalError when the status parameter is a string. #### Steps/Code to Reproduce ```python import bottle # This should work but now raises UnboundLocalError res...
bottlepy__bottle.a8dfef30.func_pm_remove_cond__wtiszsqd
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/bottlepy__bottle.a8dfef30/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Status line validation removed causing UnboundLocalError When setting response status, the validation lo...
26
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/bottlepy__bottle.a8dfef30.func_pm_remove_cond__wtiszsqd.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/bottle.py b/bottle.py index 8f6a686..dc7a207 100755 --- a/bottle.py +++ b/bottle.py @@ -1694,6 +1694,8 @@ class BaseResponse(object): def _set_status(self, status): if isinstance(status, int): code, status = status, _HTTP_STATUS_LINES.get(status) + else: + code ...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/test_environ.py::TestResponse::test_clone_cookie", "test/test_environ.py::TestResponse::test_constructor_status", "test/test_environ.py::TestR...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
344,984
3,469
7,396
eog-gemini-gemini-3.5-flash-oh-task_20251208_123126_960_c4ad7a70_f974dae7
null
null
null
null
null
Andrew Bell from Avery Ltd reached out through a social channel reporting that their installed hardware 'palo alto networks pa-3220 variant 8' is producing unusually high operating noise. Please log a new support case for this issue and register an interaction.
task_20251208_123126_960_c4ad7a70_f974dae7
csm
CSM Agent Policy You are a Customer Service Management assistant. Your goal is to assist users in the Customer Service Management lifecycle by helping them register cases, validate entitlements, manage customer assets, raise escalations, attach relevant knowledge, close cases, and in other related processes effectivel...
Andrew Bell from Avery Ltd reached out through a social channel reporting that their installed hardware 'palo alto networks pa-3220 variant 8' is producing unusually high operating noise. Please log a new support case for this issue and register an interaction.
[ "find_account", "retrieve_installed_products", "find_user", "register_new_interaction", "find_contacts", "create_new_case", "find_product" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-csm-server", "mcp_server_url": "http://localhost:8001", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/csm/dbs/db_1762254390925_u5icw4thh.sql", "context": {"x-user-email": "sara.clark93@servicenow.com"}, "user_info": ""}]
[{"verifier_type": "database_state", "name": "Case Successfully Created", "description": "Confirms that a new moderate-priority, social-channel case was created for Avery Ltd with correct contact, product, installed product linkage.", "gym_name": "sn-csm-server", "validation_config": {"query": "SELECT COUNT(*) FROM cus...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"csm","passed_verifiers":2,"resolved":true,"score":1,"total_verifiers":2,"verification_results":{"Case Successfully Created":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) FROM customer_case WHE...
null
Andrew Bell from Avery Ltd reached out through a social channel reporting that their installed hardware 'palo alto networks pa-3220 variant 8' is producing unusually high operating noise. Please log a new support case for this issue and register an interaction.
10
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-csm_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251208_123126_960_c4ad7a70_f974dae7.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gemini/gemini-3.5-flash
null
null
71,145
3,614
4,532
eog-gemini-gemini-3.5-flash-oh-task_20260109_074002_906_060e2c81_e0a9e99d
null
null
null
null
null
Locate the latest non-trashed file contains 'Budget'. List its revision history so I can see every saved version. Fetch the most recent revision’s metadata (who changed it, when, and size), and mark that revision to keep forever so it can’t be pruned. Then export this frozen revision as a PDF for audit records and add ...
task_20260109_074002_906_060e2c81_e0a9e99d
drive
# **Drive Management Assistant Policy** **Role:** Drive Management Assistant **Mandate:** Secure and accurate management of user content, permissions, and organization within Google Drive and Shared Drives. You must operate exclusively based on **confirmed user permissions**, **existing file states**, and **database...
Locate the latest non-trashed file contains 'Budget'. List its revision history so I can see every saved version. Fetch the most recent revision’s metadata (who changed it, when, and size), and mark that revision to keep forever so it can’t be pruned. Then export this frozen revision as a PDF for audit records and add ...
[ "get_revisions", "list_files", "update_revisions", "list_revisions", "update_files", "export_files" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-google-drive-mcp", "mcp_server_url": "http://localhost:8009", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/drive/dbs/db_1764776317015_jplw4hf1v.sql", "context": {"X-Auth-Token": "ya29.JYFh8MFYw3YZG9AI8AZyFL3NhGR3QWTE_3DEf7tfuylMyHGDeZHrhq4bXs0puqbBbC8YMtimpmDcmEurRJ0IYYQYwVnueud...
[{"verifier_type": "database_state", "name": "Correct File Selected", "description": "Verify that the agent selected the most recently modified, non-trashed file named 'Project Phoenix Spec'.", "gym_name": "gym-google-drive-mcp", "validation_config": {"query": "WITH candidates AS (SELECT id FROM files WHERE name like '...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"drive","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":3,"verification_results":{"Correct File Selected":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"WITH candidates AS (SELECT id FROM file...
null
Locate the latest non-trashed file contains 'Budget'. List its revision history so I can see every saved version. Fetch the most recent revision’s metadata (who changed it, when, and size), and mark that revision to keep forever so it can’t be pruned. Then export this frozen revision as a PDF for audit records and add ...
9
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-drive_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260109_074002_906_060e2c81_e0a9e99d.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
261,431
5,063
5,928
swes-gemini-gemini-3.5-flash-oh-pydantic__pydantic.acb0f10f.func_pm_ctrl_invert_if__egt6yabo
diff --git a/pydantic/_internal/_generate_schema.py b/pydantic/_internal/_generate_schema.py index 9ec8a61..0dde465 100644 --- a/pydantic/_internal/_generate_schema.py +++ b/pydantic/_internal/_generate_schema.py @@ -2273,15 +2273,15 @@ class GenerateSchema: return_schema = self.generate_schema(return_...
[ "tests/test_docs.py::test_docstrings_examples[pydantic/functional_serializers.py:364-382]", "tests/test_docs.py::test_docs_examples[docs/concepts/serialization.md:190-225]", "tests/test_docs.py::test_docs_examples[docs/concepts/serialization.md:295-309]", "tests/test_parse.py::test_model_validate_root", "te...
[ "tests/benchmarks/test_attribute_access.py::test_setattr", "tests/benchmarks/test_attribute_access.py::test_getattr", "tests/benchmarks/test_discriminated_unions.py::test_schema_build", "tests/benchmarks/test_discriminated_unions.py::test_efficiency_with_highly_nested_examples", "tests/benchmarks/test_fasta...
swebench/swesmith.x86_64.pydantic_1776_pydantic.acb0f10f
swesmith/pydantic__pydantic.acb0f10f
Model serializer mode logic inverted causing incorrect serialization behavior #### Description When using `@model_serializer` with `mode='wrap'`, the serializer is incorrectly treated as a plain serializer, and vice versa. This causes unexpected serialization behavior where wrap serializers don't receive the handler ...
pydantic__pydantic.acb0f10f.func_pm_ctrl_invert_if__egt6yabo
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pydantic__pydantic.acb0f10f/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Model serializer mode logic inverted causing incorrect serialization behavior #### Description When u...
57
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pydantic__pydantic.acb0f10f.func_pm_ctrl_invert_if__egt6yabo.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/pydantic/_internal/_generate_schema.py b/pydantic/_internal/_generate_schema.py index 0dde465..8b5520a 100644 --- a/pydantic/_internal/_generate_schema.py +++ b/pydantic/_internal/_generate_schema.py @@ -2273,15 +2273,15 @@ class GenerateSchema: return_schema = self.generate_schema(return_...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_docs.py::test_docstrings_examples[pydantic/functional_serializers.py:364-382]", "tests/test_docs.py::test_docs_examples[docs/concepts/serialization.md:19...
+ cd /testbed + : '>>>>> Start Test Output' + /root/.local/bin/uv run pytest --disable-warnings --color=no --tb=no --verbose Building pydantic @ file:///testbed Built pydantic @ file:///testbed Uninstalled 1 package in 1ms Installed 1 package in 8ms ============================= test session starts ===========...
1,744,356
10,864
17,793
swes-gpt-5.5-2026-04-23-oh-bottlepy__bottle.a8dfef30.func_pm_ctrl_invert_if__ki1zfode
diff --git a/bottle.py b/bottle.py index 286b5bc..5d929e7 100755 --- a/bottle.py +++ b/bottle.py @@ -4410,20 +4410,19 @@ class StplParser(object): code_line = _blk2 self.indent_mod -= 1 elif _cend: # The end-code-block template token (usually '%>') - if mul...
[ "test/test_auth.py::TestBasicAuth::test__header", "test/test_exc.py::TestAppException::test_noncatched_error", "test/test_exc.py::TestAppException::test_other_error", "test/test_mount.py::TestAppMounting::test_mount", "test/test_mount.py::TestAppMounting::test_no_slash_prefix", "test/test_outputfilter.py:...
[ "test/test_app.py::TestApplicationObject::test_setattr", "test/test_config.py::TestConfDict::test_gc_overlays", "test/test_config.py::TestConfDict::test_isadict", "test/test_config.py::TestConfDict::test_load_dict", "test/test_config.py::TestConfDict::test_load_module", "test/test_config.py::TestConfDict:...
swebench/swesmith.x86_64.bottlepy_1776_bottle.a8dfef30
swesmith/bottlepy__bottle.a8dfef30
Template parsing broken for multiline code blocks with end-of-block markers Description When using multiline code blocks in templates with the `<% ... %>` syntax, the parser incorrectly handles the end-of-block marker (`%>`), causing syntax errors and incorrect code generation. Steps/Code to Reproduce ```python fro...
bottlepy__bottle.a8dfef30.func_pm_ctrl_invert_if__ki1zfode
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/bottlepy__bottle.a8dfef30/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Template parsing broken for multiline code blocks with end-of-block markers Description When using mult...
40
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/bottlepy__bottle.a8dfef30.func_pm_ctrl_invert_if__ki1zfode.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/bottle.py b/bottle.py index 5d929e7..5fb58ec 100755 --- a/bottle.py +++ b/bottle.py @@ -4410,17 +4410,17 @@ class StplParser(object): code_line = _blk2 self.indent_mod -= 1 elif _cend: # The end-code-block template token (usually '%>') - if mul...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/test_auth.py::TestBasicAuth::test__header", "test/test_exc.py::TestAppException::test_noncatched_error", "test/test_exc.py::TestAppException::t...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,284,623
2,907
10,655
eog-gpt-5.5-2026-04-23-oh-task_20260105_074654_229_bca89e7d_39d9c8ac
null
null
null
null
null
Create a calendar named 'Hackathon 2026' in the UTC time, ensuring it uses a color ID clearly different from all existing calendars for easy identification. Schedule a 48-hour all-day event titled 'Build, Ship, Win!' from March 20 to March 22. Also, schedule a Google Meet 'Hackathon Kickoff Session' at 09:00 UTC on Mar...
task_20260105_074654_229_bca89e7d_39d9c8ac
calendar
You are a Google Calendar automation agent with full administrative permissions to manage users, meetings, recordings, and messages. Operate in a safe and fully authorized environment — you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform...
Create a calendar named 'Hackathon 2026' in the UTC time, ensuring it uses a color ID clearly different from all existing calendars for easy identification. Schedule a 48-hour all-day event titled 'Build, Ship, Win!' from March 20 to March 22. Also, schedule a Google Meet 'Hackathon Kickoff Session' at 09:00 UTC on Mar...
[ "list_events", "create_calendar", "add_calendar_to_list", "create_event", "create_event", "get_colors", "get_calendar_list", "watch_events", "create_event" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1762868439331_kf914hbmw.sql", "context": {"x-access-token": "ya29.A0ARrdaM-k9Vq7GzY2pL4mQf8sN1xT0bR3uHcJWv5yKzP6eF2.qwErTyUIopASDfGhJkLzXcVbNm12_34-56"}, "user_info...
[{"verifier_type": "database_state", "name": "Calendar Created", "description": "Check calendar creation with unique color", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) AS count FROM calendars WHERE user_id = 'alice_manager' AND summary = 'Hackathon 2026' AND time_zone = 'UTC' AND color_...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":6,"resolved":true,"score":1,"total_verifiers":6,"verification_results":{"Alert Setup":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) AS count FROM watch_channels WH...
null
Create a calendar named 'Hackathon 2026' in the UTC time, ensuring it uses a color ID clearly different from all existing calendars for easy identification. Schedule a 48-hour all-day event titled 'Build, Ship, Win!' from March 20 to March 22. Also, schedule a Google Meet 'Hackathon Kickoff Session' at 09:00 UTC on Mar...
11
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260105_074654_229_bca89e7d_39d9c8ac.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gpt-5.5-2026-04-23
null
null
49,487
2,496
3,618
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-lincolnloop__python-qrcode.456b01d4.func_pm_class_rm_funcs__i2sijd4c
diff --git a/qrcode/image/pil.py b/qrcode/image/pil.py index 57ee13a..67c3925 100644 --- a/qrcode/image/pil.py +++ b/qrcode/image/pil.py @@ -9,40 +9,6 @@ class PilImage(qrcode.image.base.BaseImage): kind = "PNG" - def new_image(self, **kwargs): - if not Image: - raise ImportError("PIL libr...
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode_pil.py::test_render_pil", "qrcode/tests/test_qrcode_pil.py::test_render_pil_background[TransParent]", "qrcode/tests/test_qrcode_pil.py::test_render_pil_background[red]", "qrcode/tests/test_qrcode_pil.py::test_render_pil_background[back_...
[ "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qrcode.py::test_add_qrdata", "qrcode/tests/test_...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
PilImage.new_image method missing causes AttributeError when creating QR codes When trying to create QR codes using the PIL image factory, I'm getting an AttributeError because the `new_image` method is missing from the PilImage class. #### Steps/Code to Reproduce ```python import qrcode # Basic QR code creation fa...
lincolnloop__python-qrcode.456b01d4.func_pm_class_rm_funcs__i2sijd4c
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> PilImage.new_image method missing causes AttributeError when creating QR codes When trying to ...
32
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.func_pm_class_rm_funcs__i2sijd4c.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/qrcode/image/pil.py b/qrcode/image/pil.py index 67c3925..efb976f 100644 --- a/qrcode/image/pil.py +++ b/qrcode/image/pil.py @@ -9,6 +9,40 @@ class PilImage(qrcode.image.base.BaseImage): kind = "PNG" + def new_image(self, **kwargs): + if not Image: + raise ImportError("PIL libr...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode_pil.py::test_render_pil", "qrcode/tests/test_qrcode_pil.py::test_render_pil_b...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
827,852
8,162
13,307
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20251212_162948_602_8ce75c22_8b25a027
null
null
null
null
null
Security cleanup time. Check my delegates - the one that's still pending verification needs to go, but keep my two verified ones. Look at my filters and delete whichever one has a "remove labels" action, not the one that adds labels. Find my subscription renewal thread and mark it important. I have two unused labels to...
task_20251212_162948_602_8ce75c22_8b25a027
email
# **Email Management Assistant Policy** **Role:** Email Management Assistant **Mandate:** Secure, accurate, and efficient management of user email communications (Messages, Drafts, Threads, Labels, Filters, and Settings). You must act solely based on **confirmed user context** and **existing data integrity rules**. ...
Security cleanup time. Check my delegates - the one that's still pending verification needs to go, but keep my two verified ones. Look at my filters and delete whichever one has a "remove labels" action, not the one that adds labels. Find my subscription renewal thread and mark it important. I have two unused labels to...
[ "email_list_delegates", "list_threads", "list_filters", "delete_label", "modify_thread", "list_labels", "delete_filter", "email_delete_delegate", "update_language_settings" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/email/dbs/db_1764696098233_cvtva4yqg.sql", "context": {"x-email-user-token": "lxu4KnpA68HYxHXE9JuwpmF-n1bnz9tojT7KiIFfN_8", "user_id": "5c402cc6-4141-4d0a-beab-29b5f94fa715", "pri...
[{"verifier_type": "database_state", "name": "Verify Label-Removing Filter Deleted", "description": "Confirms the filter with removeLabelIds action was removed", "gym_name": "gym-email-mcp", "validation_config": {"query": "SELECT COUNT(*) FROM filters WHERE id = 'filter_69cef1e1-f5d4-4d3f-a0e4-344d133d7182'", "expected...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"email","passed_verifiers":6,"resolved":true,"score":1,"total_verifiers":6,"verification_results":{"Verify Follow-up Label Deleted":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 0","expected":0,"passed":true,"query":"SELECT COUNT(*) FROM labels WHE...
null
Security cleanup time. Check my delegates - the one that's still pending verification needs to go, but keep my two verified ones. Look at my filters and delete whichever one has a "remove labels" action, not the one that adds labels. Find my subscription renewal thread and mark it important. I have two unused labels to...
11
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-email_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251212_162948_602_8ce75c22_8b25a027.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
23,735
930
2,092
swes-gemini-gemini-3.5-flash-oh-lincolnloop__python-qrcode.456b01d4.func_pm_remove_cond__72ttrvkx
diff --git a/qrcode/main.py b/qrcode/main.py index 46116b5..8344e49 100644 --- a/qrcode/main.py +++ b/qrcode/main.py @@ -293,19 +293,12 @@ class QRCode(Generic[GenericImage]): :param tty: use fixed TTY color codes (forces invert=True) :param invert: invert the ASCII characters (solid <-> transparent) ...
[ "qrcode/tests/test_qrcode.py::test_print_ascii_stdout", "qrcode/tests/test_qrcode.py::test_print_ascii" ]
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qr...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
print_ascii method broken after recent changes #### Description The `print_ascii` method is not working correctly. When calling it without arguments, it raises an AttributeError, and when providing an output stream, the output format is incorrect. #### Steps/Code to Reproduce ```python import qrcode import io # Th...
lincolnloop__python-qrcode.456b01d4.func_pm_remove_cond__72ttrvkx
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> print_ascii method broken after recent changes #### Description The `print_ascii` method is n...
58
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.func_pm_remove_cond__72ttrvkx.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/qrcode/main.py b/qrcode/main.py index 8344e49..a9c08b0 100644 --- a/qrcode/main.py +++ b/qrcode/main.py @@ -293,12 +293,19 @@ class QRCode(Generic[GenericImage]): :param tty: use fixed TTY color codes (forces invert=True) :param invert: invert the ASCII characters (solid <-> transparent) ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_qrcode.py::test_print_ascii_stdout", "qrcode/tests/test_qrcode.py::test_print_ascii" ], "failure": [] }, "...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,639,696
24,124
34,961
eog-gpt-5.5-2026-04-23-oh-task_20260109_172559_279_e6427839_f711406f
null
null
null
null
null
Please stand up a new lifecycle event for our Global Assignment Offboarding Program in the ServiceNow organization and ensure the configuration is complete end to end. Start by creating a topic category called Global Assignments under the Workforce Administration COE, then add three topic details beneath it: Assignment...
task_20260109_172559_279_e6427839_f711406f
hr
HR MANAGEMENT ASSISTANT POLICY Role: HR Management Assistant System Scope: Internal Employee Services, HR Case Management, Personnel Data, and Policy Fulfillment. Compliance Level: Strict, especially regarding PII/PHI. You are an HR Management Assistant. Your primary goal is to facilitate the efficient, secure, and...
Please stand up a new lifecycle event for our Global Assignment Offboarding Program in the ServiceNow organization and ensure the configuration is complete end to end. Start by creating a topic category called Global Assignments under the Workforce Administration COE, then add three topic details beneath it: Assignment...
[ "create_new_service_template", "find_group_by_name", "list_user_groups", "create_new_hr_service", "list_group_members", "add_new_topic_detail", "register_hr_criteria", "get_topic_categories", "add_new_topic_category", "get_service_template_by_name" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-hr-internal", "mcp_server_url": "http://localhost:8008", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hr/dbs/db_1765227945688_occ0a7ed3.sql", "context": {"x-hr-user-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJVU0VSXzAzOSIsImlzcyI6Iml0c20ifQ.Ab6_dflKxwRJSMeKKF2QT4fwpM...
[{"verifier_type": "database_state", "name": "Topic Category created: Global Assignments under Workforce Administration COE", "description": "Topic Category created: Global Assignments under Workforce Administration COE", "gym_name": "sn-hr-internal", "validation_config": {"query": "SELECT COUNT(*) AS cnt\nFROM topic_c...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hr","passed_verifiers":0,"resolved":false,"score":0,"total_verifiers":6,"verification_results":{"HR Criteria (approval): Global Assignment Approval limited to a Service Desk group":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 1","expected":1,"pas...
null
Please stand up a new lifecycle event for our Global Assignment Offboarding Program in the ServiceNow organization and ensure the configuration is complete end to end. Start by creating a topic category called Global Assignments under the Workforce Administration COE, then add three topic details beneath it: Assignment...
1
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hr_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260109_172559_279_e6427839_f711406f.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
5,775
635
759
swes-gpt-5.5-2026-04-23-oh-lincolnloop__python-qrcode.456b01d4.combine_file__9kz9zzgb
diff --git a/qrcode/image/styles/colormasks.py b/qrcode/image/styles/colormasks.py index 43186b2..58a919d 100644 --- a/qrcode/image/styles/colormasks.py +++ b/qrcode/image/styles/colormasks.py @@ -25,7 +25,7 @@ class QRColorMask: paint_color = back_color def initialize(self, styledPilImage, image): - ...
[ "qrcode/tests/test_qrcode_pil.py::test_render_styled_Image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image_path", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_drawer[CircleModuleDrawer]", "qrcode...
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qr...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
QRColorMask initialization and background pixel retrieval broken #### Description When using styled QR code images with color masks, the color mask initialization is broken and background pixel retrieval returns incorrect colors. This affects all styled QR code generation with custom color masks and module drawers. ...
lincolnloop__python-qrcode.456b01d4.combine_file__9kz9zzgb
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> QRColorMask initialization and background pixel retrieval broken #### Description When using ...
35
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.combine_file__9kz9zzgb.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/qrcode/image/styles/colormasks.py b/qrcode/image/styles/colormasks.py index 58a919d..afe1c35 100644 --- a/qrcode/image/styles/colormasks.py +++ b/qrcode/image/styles/colormasks.py @@ -25,7 +25,7 @@ class QRColorMask: paint_color = back_color def initialize(self, styledPilImage, image): - ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_qrcode_pil.py::test_render_styled_Image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image", "qrcode/tests/...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,000,034
2,223
10,061
eog-gemini-gemini-3.5-flash-oh-task_20260121_183027_058_1afb323a_c53cb400
null
null
null
null
null
Could you please onboard a new support agent, Susan Adams (susan.adams075@servicenow.com)? She's joining the Customer Service team this week and will be taking over several open customer cases. Please create her user profile as active agent, and assign her location at Plot number 685, 'Ferguson Neck' Street, Mumbai, In...
task_20260121_183027_058_1afb323a_c53cb400
csm
# **CSM Assistant Policy** You are a Customer Service Management administrator. Your goal is to assist users in the Customer Service Management lifecycle by helping them register cases, validate entitlements, manage customer assets, raise escalations, attach relevant knowledge, close cases and in other related process...
Could you please onboard a new support agent, Susan Adams (susan.adams075@servicenow.com)? She's joining the Customer Service team this week and will be taking over several open customer cases. Please create her user profile as active agent, and assign her location at Plot number 685, 'Ferguson Neck' Street, Mumbai, In...
[ "find_locations", "find_user", "list_users", "assign_case_to_user", "get_cases_assigned_to", "update_user_details", "add_new_user", "add_new_group_member", "add_location" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-csm-server", "mcp_server_url": "http://localhost:8001", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/csm/dbs/db_1762254390925_u5icw4thh.sql", "context": {"x-user-email": "jose.roberson@servicenow.com"}, "user_info": {"user_id": 32, "name": "Jose Roberson", "email": "jose.roberson...
[{"verifier_type": "database_state", "name": "New Agent Created", "description": "Confirms that Susan Adams has been created and activated as an agent.", "gym_name": "sn-csm-server", "validation_config": {"query": "SELECT COUNT(*) FROM user WHERE first_name = 'Susan' AND last_name = 'Adams' AND email = 'susan.adams075@...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"csm","passed_verifiers":5,"resolved":true,"score":1,"total_verifiers":5,"verification_results":{"Cases Reassigned Correctly":{"actual":2,"comparison_type":"equals","details":"Comparison equals: 2 vs 2","expected":2,"passed":true,"query":"SELECT COUNT(*) FROM customer_case WH...
null
Could you please onboard a new support agent, Susan Adams (susan.adams075@servicenow.com)? She's joining the Customer Service team this week and will be taking over several open customer cases. Please create her user profile as active agent, and assign her location at Plot number 685, 'Ferguson Neck' Street, Mumbai, In...
17
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-csm_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260121_183027_058_1afb323a_c53cb400.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gemini/gemini-3.5-flash
null
null
428,157
10,465
11,495
swes-gemini-gemini-3.5-flash-oh-getmoto__moto.694ce1f4.pr_5093
diff --git a/moto/mediaconnect/models.py b/moto/mediaconnect/models.py index bbb34e7..7b46be4 100644 --- a/moto/mediaconnect/models.py +++ b/moto/mediaconnect/models.py @@ -123,6 +123,12 @@ class MediaConnectBackend(BaseBackend): sources: List[Dict[str, Any]], vpc_interfaces: List[Dict[str, Any]], ...
[ "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_alternative_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_list_flows_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_describe_flow_succeeds...
[ "tests/test_mediaconnect/test_mediaconnect.py::test_unknown_flow", "tests/test_mediaconnect/test_mediaconnect.py::test_tag_resource_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_add_flow_vpc_interfaces_fails", "tests/test_mediaconnect/test_mediaconnect.py::test_remove_flow_vpc_interface_fails...
swebench/swesmith.x86_64.getmoto_1776_moto.694ce1f4
swesmith/getmoto__moto.694ce1f4
Flow Creation and Update Issues in MediaConnect Description After the recent changes, there seems to be an issue with creating and updating flows in the MediaConnect service. When attempting to create a flow, the expected outputs and source configurations are not being set correctly. Specifically, the `ListenerAddres...
getmoto__moto.694ce1f4.pr_5093
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/getmoto__moto.694ce1f4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Flow Creation and Update Issues in MediaConnect Description After the recent changes, there seems to be an...
47
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/getmoto__moto.694ce1f4.pr_5093.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/AGENTS.md b/AGENTS.md new file mode 100644 index 0000000..bce8b97 --- /dev/null +++ b/AGENTS.md @@ -0,0 +1,6 @@ +# Moto - MediaConnect Service Insights + +## Learnings & Patterns +- **Parameter validation & Schema translation:** Under real boto3 calls, client-side parameter validation is performed based on...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_list_flows_succeeds" ], ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,847,059
21,143
31,349
swes-gemini-gemini-3.5-flash-oh-kurtmckee__feedparser.cad965a3.lm_rewrite__e93hc3ot
diff --git a/feedparser/namespaces/mediarss.py b/feedparser/namespaces/mediarss.py index b057580..abbb491 100644 --- a/feedparser/namespaces/mediarss.py +++ b/feedparser/namespaces/mediarss.py @@ -99,10 +99,7 @@ class Namespace: restriction = self.pop("restriction") if restriction is not None and rest...
[ "tests/test_well_formed.py::test_strict_parser[info1147]", "tests/test_well_formed.py::test_loose_parser[info1147]" ]
[ "tests/test_temporary_behavior.py::test_issue_328_fallback_behavior", "tests/test_make_safe_absolute_uri.py::test_catch_value_error[http://bad]test/]", "tests/test_feedparser_dict.py::test_single_mapping_target_2", "tests/test_feedparser_dict.py::test_license", "tests/test_feedparser_dict.py::test_category"...
swebench/swesmith.x86_64.kurtmckee_1776_feedparser.cad965a3
swesmith/kurtmckee__feedparser.cad965a3
Media RSS restriction content is not properly parsed I'm working with RSS feeds that contain Media RSS elements, specifically the `<media:restriction>` element. I've noticed that the content of this element is not being properly parsed when it contains multiple country codes. Here's a simple example that demonstrates...
kurtmckee__feedparser.cad965a3.lm_rewrite__e93hc3ot
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/kurtmckee__feedparser.cad965a3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Media RSS restriction content is not properly parsed I'm working with RSS feeds that contain Media ...
51
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/kurtmckee__feedparser.cad965a3.lm_rewrite__e93hc3ot.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/AGENTS.md b/AGENTS.md new file mode 100644 index 0000000..5f29e2d --- /dev/null +++ b/AGENTS.md @@ -0,0 +1,8 @@ +# feedparser agent context + +## Media RSS Parsing +- Location of namespace handlers: `feedparser/namespaces/` +- Media RSS specific logic is implemented in `feedparser/namespaces/mediarss.py`. ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_well_formed.py::test_strict_parser[info1147]", "tests/test_well_formed.py::test_loose_parser[info1147]" ], "failure": [] ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,181,205
10,519
20,083
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20251218_193035_592_391f1eff_5293f595
null
null
null
null
null
This incident INC-000068 has been in the L2 Field Support team queue for some time now awaiting action from the caller. The laptop reported as not powering on is currently on hold until the caller brings in the device for further hardware checks, after which the required parts can be identified and arranged. Notify the...
task_20251218_193035_592_391f1eff_5293f595
itsm
# **ITSM Assistant Policy** **Role:** ITSM Assistant (IT Service Management) **Mandate:** Ensure the efficient, secure, and controlled delivery of IT services by managing the full lifecycle of Incidents, Service Requests, Changes, Problems, and Configuration Items (CIs) in strict accordance with defined policies and ...
This incident INC-000068 has been in the L2 Field Support team queue for some time now awaiting action from the caller. The laptop reported as not powering on is currently on hold until the caller brings in the device for further hardware checks, after which the required parts can be identified and arranged. Notify the...
[ "send_notification", "get_user", "update_incident", "find_incident_by_number" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-itsm-mcp", "mcp_server_url": "http://localhost:8006", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/itsm/dbs/db_1765893060767_9rwtfbbp7.sql", "context": {"x-itsm-user-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJVU0VSXzAzNyIsImlzcyI6Iml0c20ifQ.Wx2_dflKxwRJSMeKKF2QT4fw...
[{"verifier_type": "database_state", "name": "Modify incident", "description": "validate incident update", "gym_name": "gym-itsm-mcp", "validation_config": {"query": "SELECT COUNT(*) FROM incident WHERE incident_id = \"INC_068\" AND urgency = \"high\" AND priority = \"critical\";", "expected_value": 1, "comparison_type...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"itsm","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":4,"verification_results":{"Modify incident":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 1","expected":1,"passed":false,"query":"SELECT COUNT(*) FROM incident WHERE incident_...
null
This incident INC-000068 has been in the L2 Field Support team queue for some time now awaiting action from the caller. The laptop reported as not powering on is currently on hold until the caller brings in the device for further hardware checks, after which the required parts can be identified and arranged. Notify the...
5
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-itsm_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251218_193035_592_391f1eff_5293f595.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
24,869
1,528
2,542
swes-gemini-gemini-3.5-flash-oh-andialbrecht__sqlparse.e57923b3.func_basic__ztx58hlb
diff --git a/sqlparse/filters/output.py b/sqlparse/filters/output.py index 253537e..e4081ed 100644 --- a/sqlparse/filters/output.py +++ b/sqlparse/filters/output.py @@ -20,13 +20,13 @@ class OutputFilter: def process(self, stmt): self.count += 1 - if self.count > 1: - varname = '{f.var...
[ "tests/test_format.py::TestOutputFormat::test_python", "tests/test_format.py::TestOutputFormat::test_python_multiple_statements", "tests/test_format.py::TestOutputFormat::test_php" ]
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_valid_args", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::t...
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
OutputFilter generates incorrect variable names for first statement #### Description When using sqlparse's output formatting with `output_format='python'` or `output_format='php'`, the first SQL statement gets an incorrect variable name. Instead of the expected base variable name (like `sql` or `$sql`), it generates ...
andialbrecht__sqlparse.e57923b3.func_basic__ztx58hlb
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> OutputFilter generates incorrect variable names for first statement #### Description When using s...
58
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_basic__ztx58hlb.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/filters/output.py b/sqlparse/filters/output.py index e4081ed..253537e 100644 --- a/sqlparse/filters/output.py +++ b/sqlparse/filters/output.py @@ -20,13 +20,13 @@ class OutputFilter: def process(self, stmt): self.count += 1 - if self.count >= 1: - varname = '{f.va...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_format.py::TestOutputFormat::test_python", "tests/test_format.py::TestOutputFormat::test_python_multiple_statements", "tests/test_format....
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,052,852
12,292
20,058
eog-gpt-5.5-2026-04-23-oh-task_20260104_143530_440_6f5b66fc_b6b24e96
null
null
null
null
null
Check my primary availability on December 10, 2025, between 9:00 AM and 5:00 PM UTC. If I’m free, move my office events to that date, highlight them in red, add carol.white@techcorp.com as an attendee, and enable monitoring to watch for any changes using https://calendar-changes/office-event-monitor-dec10, as this is a...
task_20260104_143530_440_6f5b66fc_b6b24e96
calendar
You are a Google Calendar automation agent with full administrative permissions to manage users, meetings, recordings, and messages. Operate in a safe and fully authorized environment — you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform...
Check my primary availability on December 10, 2025, between 9:00 AM and 5:00 PM UTC. If I’m free, move my office events to that date, highlight them in red, add carol.white@techcorp.com as an attendee, and enable monitoring to watch for any changes using https://calendar-changes/office-event-monitor-dec10, as this is a...
[ "watch_events", "patch_event", "get_calendar_from_list", "get_colors", "query_freebusy", "list_events" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1764361046464_47vxzhsli.sql", "context": {"x-access-token": "ya29.A0ARrdaM-Zx8Nw3Q4pVb6Ls9R1mT0cG2uF5yH7kJd8sA1Lq2.wErtYuIoPaSdFgHjKlZxCvBnM987_65-43"}, "user_info"...
[{"verifier_type": "database_state", "name": "event verifier", "description": "event verifier", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) FROM events WHERE summary = 'Office Day - Collaboration Sessions' AND color_id = 11 AND visibility = 'confidential' AND start_datetime = '2025-12-10...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":3,"resolved":true,"score":1,"total_verifiers":3,"verification_results":{"attendees verifier":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT count(*) FROM attendees WHERE eve...
null
Check my primary availability on December 10, 2025, between 9:00 AM and 5:00 PM UTC. If I’m free, move my office events to that date, highlight them in red, add carol.white@techcorp.com as an attendee, and enable monitoring to watch for any changes using https://calendar-changes/office-event-monitor-dec10, as this is a...
9
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260104_143530_440_6f5b66fc_b6b24e96.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gpt-5.5-2026-04-23
null
null
72,048
1,638
2,512
eog-gemini-gemini-3.5-flash-oh-task_20251201_164641_724_109c89c0_1b29acb9
null
null
null
null
null
Globex has informed us that their current primary contact has resigned and should no longer appear on any of their open support items, so please ensure that this user is completely removed from all open cases by reassigning any open cases to the most recent secondary contact on the account, promote that secondary conta...
task_20251201_164641_724_109c89c0_1b29acb9
csm
CSM Agent Policy You are a Customer Service Management assistant. Your goal is to assist users in the Customer Service Management lifecycle by helping them register cases, validate entitlements, manage customer assets, raise escalations, attach relevant knowledge, close cases, and in other related processes effectivel...
Globex has informed us that their current primary contact has resigned and should no longer appear on any of their open support items, so please ensure that this user is completely removed from all open cases by reassigning any open cases to the most recent secondary contact on the account, promote that secondary conta...
[ "update_user_details", "update_case", "find_contacts", "list_users", "send_notification", "update_contact", "find_account", "search_cases", "create_new_case" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-csm-server", "mcp_server_url": "http://localhost:8001", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/csm/dbs/db_1762254390925_u5icw4thh.sql", "context": {"x-user-email": "jose.roberson@servicenow.com"}, "user_info": {"user_id": 32, "name": "Jose Roberson", "email": "jose.roberson...
[{"verifier_type": "database_state", "name": "Move all open cases", "description": "All Open Cases moved from the resigned primary contact to the updated primary contact", "gym_name": "sn-csm-server", "validation_config": {"query": "SELECT COUNT(*) AS cnt\nFROM customer_case\nWHERE account_id = 4\n AND contact_id = 2\...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"csm","passed_verifiers":32,"resolved":false,"score":0,"total_verifiers":44,"verification_results":{"Move all open cases":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 0","expected":0,"passed":true,"query":"SELECT COUNT(*) AS cnt\nFROM customer_cas...
null
Globex has informed us that their current primary contact has resigned and should no longer appear on any of their open support items, so please ensure that this user is completely removed from all open cases by reassigning any open cases to the most recent secondary contact on the account, promote that secondary conta...
19
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-csm_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251201_164641_724_109c89c0_1b29acb9.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
224,066
4,389
5,930
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-lincolnloop__python-qrcode.456b01d4.func_pm_remove_cond__72ttrvkx
diff --git a/qrcode/main.py b/qrcode/main.py index 46116b5..8344e49 100644 --- a/qrcode/main.py +++ b/qrcode/main.py @@ -293,19 +293,12 @@ class QRCode(Generic[GenericImage]): :param tty: use fixed TTY color codes (forces invert=True) :param invert: invert the ASCII characters (solid <-> transparent) ...
[ "qrcode/tests/test_qrcode.py::test_print_ascii_stdout", "qrcode/tests/test_qrcode.py::test_print_ascii" ]
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qr...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
print_ascii method broken after recent changes #### Description The `print_ascii` method is not working correctly. When calling it without arguments, it raises an AttributeError, and when providing an output stream, the output format is incorrect. #### Steps/Code to Reproduce ```python import qrcode import io # Th...
lincolnloop__python-qrcode.456b01d4.func_pm_remove_cond__72ttrvkx
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> print_ascii method broken after recent changes #### Description The `print_ascii` method is n...
65
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.func_pm_remove_cond__72ttrvkx.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/qrcode/main.py b/qrcode/main.py index 8344e49..a9c08b0 100644 --- a/qrcode/main.py +++ b/qrcode/main.py @@ -293,12 +293,19 @@ class QRCode(Generic[GenericImage]): :param tty: use fixed TTY color codes (forces invert=True) :param invert: invert the ASCII characters (solid <-> transparent) ...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_qrcode.py::test_print_ascii_stdout", "qrcode/tests/test_qrcode.py::test_print_ascii" ], "failure": [] }, "...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,029,746
65,741
76,053
swes-gpt-5.5-2026-04-23-oh-andialbrecht__sqlparse.e57923b3.func_basic__7p73klsa
diff --git a/sqlparse/filters/output.py b/sqlparse/filters/output.py index 253537e..afff2f8 100644 --- a/sqlparse/filters/output.py +++ b/sqlparse/filters/output.py @@ -78,45 +78,36 @@ class OutputPHPFilter(OutputFilter): varname_prefix = '$' def _process(self, stream, varname, has_nl): - # SQL query...
[ "tests/test_format.py::TestOutputFormat::test_php" ]
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_valid_args", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::t...
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
PHP output format generates incorrect syntax #### Description The PHP output format is producing malformed PHP code with incorrect operators and formatting. #### Steps/Code to Reproduce ```python import sqlparse sql = 'select * from foo;' result = sqlparse.format(sql, output_format='php', reindent=True) print("Gen...
andialbrecht__sqlparse.e57923b3.func_basic__7p73klsa
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> PHP output format generates incorrect syntax #### Description The PHP output format is producing ...
37
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_basic__7p73klsa.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/filters/output.py b/sqlparse/filters/output.py index afff2f8..43fe932 100644 --- a/sqlparse/filters/output.py +++ b/sqlparse/filters/output.py @@ -79,11 +79,9 @@ class OutputPHPFilter(OutputFilter): def _process(self, stream, varname, has_nl): if self.count > 1: - yield ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_format.py::TestOutputFormat::test_php" ], "failure": [] }, "PASS_TO_PASS": { "success": [ "tests/test...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
792,555
3,654
11,471
eog-gpt-5.5-2026-04-23-oh-task_20251128_123301_181_64a4ebb9_4c1d5329
null
null
null
null
null
I need to securely archive the 'Technical Spec.pdf' file. First, create a new folder named 'Secure_Vault' in My Drive. Copy the 'Technical Spec.pdf' into that folder. Rename the new copy to 'Technical_Spec_SECURED.pdf'. Add a comment to this new file saying 'Secured'. Finally, delete the original 'Technical Spec.pdf' f...
task_20251128_123301_181_64a4ebb9_4c1d5329
drive
You are a helpful Drive assistant with access to all available tools. Operate in a safe and fully authorized environment—you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform exactly one lookup per entity type, verify that you are reusing ...
I need to securely archive the 'Technical Spec.pdf' file. First, create a new folder named 'Secure_Vault' in My Drive. Copy the 'Technical Spec.pdf' into that folder. Rename the new copy to 'Technical_Spec_SECURED.pdf'. Add a comment to this new file saying 'Secured'. Finally, delete the original 'Technical Spec.pdf' f...
[ "create_comments", "copy_files", "list_files", "update_files", "delete_files", "create_file" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-google-drive-mcp", "mcp_server_url": "http://localhost:8009", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/drive/dbs/db_1763988076613_f5g41taxz.sql", "context": {"x-auth-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIwMUpCUU0zSzhUR0haWDlSTlZRTTVZMlA3VyIsImVtYWlsIjoiam9...
[{"verifier_type": "database_state", "name": "Verify Folder", "description": "Checks if 'Secure_Vault' folder exists.", "gym_name": "gym-google-drive-mcp", "validation_config": {"query": "SELECT COUNT(*) FROM files WHERE name = 'Secure_Vault' AND mime_type = 'application/vnd.google-apps.folder' AND created_by = '01JBQM...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"drive","passed_verifiers":4,"resolved":true,"score":1,"total_verifiers":4,"verification_results":{"Verify Comment":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(c.id) FROM comments c JOIN files f...
null
I need to securely archive the 'Technical Spec.pdf' file. First, create a new folder named 'Secure_Vault' in My Drive. Copy the 'Technical Spec.pdf' into that folder. Rename the new copy to 'Technical_Spec_SECURED.pdf'. Add a comment to this new file saying 'Secured'. Finally, delete the original 'Technical Spec.pdf' f...
6
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-drive_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251128_123301_181_64a4ebb9_4c1d5329.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gpt-5.5-2026-04-23
null
null
25,328
511
940
swes-gemini-gemini-3.5-flash-oh-andialbrecht__sqlparse.e57923b3.func_basic__jxajkvtq
diff --git a/sqlparse/filters/reindent.py b/sqlparse/filters/reindent.py index 7dc2b82..f8125d7 100644 --- a/sqlparse/filters/reindent.py +++ b/sqlparse/filters/reindent.py @@ -42,9 +42,8 @@ class ReindentFilter: def _get_offset(self, token): raw = ''.join(map(str, self._flatten_up_to_token(token))) - ...
[ "tests/test_format.py::TestFormatReindent::test_parenthesis", "tests/test_format.py::TestFormatReindent::test_where", "tests/test_format.py::TestFormatReindent::test_identifier_list", "tests/test_format.py::TestFormatReindent::test_identifier_list_with_wrap_after", "tests/test_format.py::TestFormatReindent:...
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_valid_args", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::t...
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
SQL formatting produces incorrect indentation after reindent changes #### Description The reindent filter is producing incorrect indentation for SQL statements. The indentation appears to be calculated incorrectly, causing statements to be formatted with wrong spacing. #### Steps/Code to Reproduce ```python import ...
andialbrecht__sqlparse.e57923b3.func_basic__jxajkvtq
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> SQL formatting produces incorrect indentation after reindent changes #### Description The reinden...
48
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_basic__jxajkvtq.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/filters/reindent.py b/sqlparse/filters/reindent.py index f8125d7..7dc2b82 100644 --- a/sqlparse/filters/reindent.py +++ b/sqlparse/filters/reindent.py @@ -42,8 +42,9 @@ class ReindentFilter: def _get_offset(self, token): raw = ''.join(map(str, self._flatten_up_to_token(token))) - ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_format.py::TestFormatReindent::test_parenthesis", "tests/test_format.py::TestFormatReindent::test_where", "tests/test_format.py::TestForm...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,648,294
14,074
20,850
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-pallets__jinja.ada0a9a6.func_basic__iklzsb2u
diff --git a/src/jinja2/nativetypes.py b/src/jinja2/nativetypes.py index 71db8cc..d936f12 100644 --- a/src/jinja2/nativetypes.py +++ b/src/jinja2/nativetypes.py @@ -75,8 +75,8 @@ class NativeCodeGenerator(CodeGenerator): def _output_child_pre( self, node: nodes.Expr, frame: Frame, finalize: CodeGenerator....
[ "tests/test_async.py::test_native_async[asyncio]", "tests/test_async.py::test_native_async[trio]", "tests/test_async.py::test_native_list_async[asyncio]", "tests/test_async.py::test_native_list_async[trio]", "tests/test_nativetypes.py::test_is_defined_native_return", "tests/test_nativetypes.py::test_undef...
[ "tests/test_api.py::TestExtendedAPI::test_item_and_attribute", "tests/test_api.py::TestExtendedAPI::test_finalize", "tests/test_api.py::TestExtendedAPI::test_finalize_constant_expression", "tests/test_api.py::TestExtendedAPI::test_no_finalize_template_data", "tests/test_api.py::TestExtendedAPI::test_context...
swebench/swesmith.x86_64.pallets_1776_jinja.ada0a9a6
swesmith/pallets__jinja.ada0a9a6
NativeCodeGenerator outputs node string representation instead of finalized code #### Description When using NativeEnvironment, template expressions are incorrectly outputting the string representation of AST nodes instead of the properly finalized code. This causes templates to render node objects like `<Name 'x' ct...
pallets__jinja.ada0a9a6.func_basic__iklzsb2u
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pallets__jinja.ada0a9a6/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> NativeCodeGenerator outputs node string representation instead of finalized code #### Description When us...
45
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pallets__jinja.ada0a9a6.func_basic__iklzsb2u.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/src/jinja2/nativetypes.py b/src/jinja2/nativetypes.py index d936f12..71db8cc 100644 --- a/src/jinja2/nativetypes.py +++ b/src/jinja2/nativetypes.py @@ -75,8 +75,8 @@ class NativeCodeGenerator(CodeGenerator): def _output_child_pre( self, node: nodes.Expr, frame: Frame, finalize: CodeGenerator....
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_async.py::test_native_async[asyncio]", "tests/test_async.py::test_native_async[trio]", "tests/test_async.py::test_native_list_async[async...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
929,580
16,295
26,391
eog-gpt-5.5-2026-04-23-oh-task_20260105_101355_392_e0ac0c6d_4d94d938
null
null
null
null
null
I need help setting up calendars for my team. First, find the user with email 'bob.smith@techcorp.com' in the system. Once you have that user, check if they already have a calendar named 'Q4 Sprint Planning' - if it exists, update its timezone to 'America/Los_Angeles' and description to 'Quarter 4 sprint planning and r...
task_20260105_101355_392_e0ac0c6d_4d94d938
calendar
You are a highly capable and intelligent AI assistant designed to help users by executing tasks. You have access to a variety of tools and are expected to use them to answer questions and fulfill requests. **Your Thought Process:** 1. **Analyze the Request:** Carefully analyze the user's latest message to understand ...
I need help setting up calendars for my team. First, find the user with email 'bob.smith@techcorp.com' in the system. Once you have that user, check if they already have a calendar named 'Q4 Sprint Planning' - if it exists, update its timezone to 'America/Los_Angeles' and description to 'Quarter 4 sprint planning and r...
[ "update_calendar", "create_calendar", "get_calendar_from_list", "add_calendar_to_list", "get_user_by_email", "update_calendar_in_list", "get_calendar_list" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1767381239880_ibbsoh6gc.sql", "context": {"x-hr-user-token": "ya29.A0ARrdaM-k9Vq7GzY2pL4mQf8sN1xT0bR3uHcJWv5yKzP6eF2.qwErTyUIopASDfGhJkLzXcVbNm12_34-56"}, "user_inf...
[{"verifier_type": "database_state", "name": "User Found by Email", "description": "Verifies that the user was found using the get_user_by_email tool with the correct email address, ensuring the model doesn't assume user_id values.", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) as count F...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":1,"resolved":false,"score":0,"total_verifiers":6,"verification_results":{"Code Review Schedule Calendar Exists":{"actual":0,"comparison_type":"greater_than","details":"Comparison greater_than: 0 vs 0","expected":0,"passed":false,"query":"SELECT C...
null
I need help setting up calendars for my team. First, find the user with email 'bob.smith@techcorp.com' in the system. Once you have that user, check if they already have a calendar named 'Q4 Sprint Planning' - if it exists, update its timezone to 'America/Los_Angeles' and description to 'Quarter 4 sprint planning and r...
3
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260105_101355_392_e0ac0c6d_4d94d938.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
15,154
971
1,248
swes-gemini-gemini-3.5-flash-oh-pydantic__pydantic.acb0f10f.func_pm_remove_cond__yd4ldncj
diff --git a/pydantic/json_schema.py b/pydantic/json_schema.py index ca031ea..b8c2330 100644 --- a/pydantic/json_schema.py +++ b/pydantic/json_schema.py @@ -470,21 +470,6 @@ class GenerateJsonSchema: """ # Generate the core-schema-type-specific bits of the schema generation: json_...
[ "tests/test_docs.py::test_docstrings_examples[pydantic/json_schema.py:2586-2620]", "tests/test_json_schema.py::test_byte_size_type", "tests/test_json_schema.py::test_ref_conflict_resolution_with_mode_difference", "tests/test_json_schema.py::test_conflicting_names", "tests/test_json_schema.py::test_plain_ser...
[ "tests/benchmarks/test_attribute_access.py::test_setattr", "tests/benchmarks/test_attribute_access.py::test_getattr", "tests/benchmarks/test_discriminated_unions.py::test_schema_build", "tests/benchmarks/test_discriminated_unions.py::test_efficiency_with_highly_nested_examples", "tests/benchmarks/test_fasta...
swebench/swesmith.x86_64.pydantic_1776_pydantic.acb0f10f
swesmith/pydantic__pydantic.acb0f10f
JSON Schema generation broken for serialization mode with PlainSerializer #### Description When generating JSON schemas in serialization mode, models with `PlainSerializer` annotations are not producing the expected schema output. The serialization schema generation appears to be completely skipped, causing the schem...
pydantic__pydantic.acb0f10f.func_pm_remove_cond__yd4ldncj
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/pydantic__pydantic.acb0f10f/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> JSON Schema generation broken for serialization mode with PlainSerializer #### Description When gener...
89
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/pydantic__pydantic.acb0f10f.func_pm_remove_cond__yd4ldncj.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/pydantic/json_schema.py b/pydantic/json_schema.py index b8c2330..660171a 100644 --- a/pydantic/json_schema.py +++ b/pydantic/json_schema.py @@ -470,6 +470,21 @@ class GenerateJsonSchema: """ # Generate the core-schema-type-specific bits of the schema generation: json_...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_json_schema.py::test_byte_size_type", "tests/test_json_schema.py::test_ref_conflict_resolution_with_mode_difference", "tests/test_json_s...
+ cd /testbed + : '>>>>> Start Test Output' + /root/.local/bin/uv run pytest --disable-warnings --color=no --tb=no --verbose Building pydantic @ file:///testbed Built pydantic @ file:///testbed Uninstalled 1 package in 0.71ms Installed 1 package in 7ms ============================= test session starts ========...
4,973,655
28,538
41,012
eog-gemini-gemini-3.5-flash-oh-task_20260106_192119_816_ad5a67e3_87a5a87c
null
null
null
null
null
Look at critical priority cases that are new for the Acme account. If you find 1 or more, they should go to the first triage group(smallest) that handles cases, but make sure that group doesn't already have too many (must be under 6 open cases). Check active enterprise support coverage for specified product in cases. C...
task_20260106_192119_816_ad5a67e3_87a5a87c
csm
CSM Agent Policy You are a Customer Service Management assistant. Your goal is to assist users in the Customer Service Management lifecycle by helping them register cases, validate entitlements, manage customer assets, raise escalations, attach relevant knowledge, close cases, and in other related processes effectivel...
Look at critical priority cases that are new for the Acme account. If you find 1 or more, they should go to the first triage group(smallest) that handles cases, but make sure that group doesn't already have too many (must be under 6 open cases). Check active enterprise support coverage for specified product in cases. C...
[ "search_cases", "update_case", "find_entitlements", "list_user_groups", "find_user_group", "find_account", "count_case_by_state", "get_accounts" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-csm-server", "mcp_server_url": "http://localhost:8001", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/csm/dbs/db_1762254390925_u5icw4thh.sql", "context": {"x-user-email": "stephanie.todd@servicenow.com"}, "user_info": {"user_id": 6, "name": "Stephanie Todd", "email": "stephanie.to...
[{"verifier_type": "database_state", "name": "Case moved to Case Triage and in_progress", "description": "Verify cases 53 are assigned to group 3, state=in_progress, with preserved assigned_to values ( 53=NULL), and no critical new cases remain for account 3", "gym_name": "sn-csm-server", "validation_config": {"query"...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"csm","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":4,"verification_results":{"Case moved to Case Triage and in_progress":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) FROM ...
null
Look at critical priority cases that are new for the Acme account. If you find 1 or more, they should go to the first triage group(smallest) that handles cases, but make sure that group doesn't already have too many (must be under 6 open cases). Check active enterprise support coverage for specified product in cases. C...
16
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-csm_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260106_192119_816_ad5a67e3_87a5a87c.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
290,094
9,623
12,052
eog-gpt-5.5-2026-04-23-oh-task_20251203_093009_456_701c5774_24b57fbd
null
null
null
null
null
Hey, can you check my vacation auto-reply? Is it active? If it is not, turn on my vacation auto-reply from Dec 20, 2025 at 9:00 AM to Jan 3, 2026 at 6:00 PM? Subject: “Out of office - On holiday” Message: “Hi, I’m on holiday during this period and will have limited access to email. I’ll reply when I’m back. For urgent...
task_20251203_093009_456_701c5774_24b57fbd
email
You are a helpful Email assistant with access to all available tools. Operate in a safe and fully authorized environment—you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform exactly one lookup per entity type, verify that you are reusing ...
Hey, can you check my vacation auto-reply? Is it active? If it is not, turn on my vacation auto-reply from Dec 20, 2025 at 9:00 AM to Jan 3, 2026 at 6:00 PM? Subject: “Out of office - On holiday” Message: “Hi, I’m on holiday during this period and will have limited access to email. I’ll reply when I’m back. For urgent...
[ "create_draft", "get_vacation_settings", "list_drafts", "update_vacation_settings" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/email/dbs/db_1763188937209_paq12p3yu.sql", "context": {"x-email-user-token": "tok_alice_2025_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6", "user_id": "user_001", "primary_email": "alice@exam...
[{"verifier_type": "database_state", "name": "Check that the vacation settings are enabled and set correctly", "description": "Check that the vacation settings are enabled and set correctly", "gym_name": "gym-email-mcp", "validation_config": {"query": "select count(*) from vacation_settings where user_id = 'user_001' a...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"email","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":4,"verification_results":{"Check that the draft is created and send as requested":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 3","expected":3,"passed":false,"query":"select...
null
Hey, can you check my vacation auto-reply? Is it active? If it is not, turn on my vacation auto-reply from Dec 20, 2025 at 9:00 AM to Jan 3, 2026 at 6:00 PM? Subject: “Out of office - On holiday” Message: “Hi, I’m on holiday during this period and will have limited access to email. I’ll reply when I’m back. For urgent...
12
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-email_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251203_093009_456_701c5774_24b57fbd.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
55,390
6,466
8,131
swes-gemini-gemini-3.5-flash-oh-iterative__dvc.1d6ea681.pr_8741
diff --git a/dvc/repo/experiments/queue/workspace.py b/dvc/repo/experiments/queue/workspace.py index 05d221b..8999822 100644 --- a/dvc/repo/experiments/queue/workspace.py +++ b/dvc/repo/experiments/queue/workspace.py @@ -13,7 +13,7 @@ from dvc.repo.experiments.exceptions import ExpQueueEmptyError from dvc.repo.experim...
[ "tests/func/experiments/test_utils.py::test_generate_random_exp_name" ]
[ "tests/func/experiments/test_queue.py::test_celery_logs[True]", "tests/func/experiments/test_queue.py::test_celery_logs[False]", "tests/func/experiments/test_queue.py::test_queue_doesnt_remove_untracked_params_file", "tests/func/experiments/test_queue.py::test_copy_paths_queue", "tests/func/experiments/test...
swebench/swesmith.x86_64.iterative_1776_dvc.1d6ea681
swesmith/iterative__dvc.1d6ea681
# Random experiment names are truncated when using DVC experiments ## Description I've discovered an issue with the random experiment name generation in DVC. When running experiments, the generated names are being truncated and only using a small subset of the available adjectives. ## To Reproduce Here's a simple s...
iterative__dvc.1d6ea681.pr_8741
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/iterative__dvc.1d6ea681/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> # Random experiment names are truncated when using DVC experiments ## Description I've discovered an issu...
40
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/iterative__dvc.1d6ea681.pr_8741.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/dvc/repo/experiments/utils.py b/dvc/repo/experiments/utils.py index 531c6ae..f3dfabb 100644 --- a/dvc/repo/experiments/utils.py +++ b/dvc/repo/experiments/utils.py @@ -283,7 +283,7 @@ def fetch_all_exps(scm: "Git", url: str, progress: Optional[Callable] = None, ** def gen_random_name(): # fmt: off ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [], "failure": [ "tests/func/experiments/test_utils.py::test_generate_random_exp_name" ] }, "PASS_TO_PASS": { "success": [ ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,281,912
26,337
36,718
eog-gemini-gemini-3.5-flash-oh-task_20260108_050214_962_e5bf1251_0cabbd3d
null
null
null
null
null
Hi, firstly, check and if not present create a forwarding email, work.backup@company.com with status as pending. Then search for the threads and tell me full thread details, especially how many emails are under the thread related to the emergency contact. Modify another thread related to new business partnerships to la...
task_20260108_050214_962_e5bf1251_0cabbd3d
email
# **Email Management Assistant Policy** **Role:** Email Management Assistant **Mandate:** Secure, accurate, and efficient management of user email communications (Messages, Drafts, Threads, Labels, Filters, and Settings). You must act solely based on **confirmed user context** and **existing data integrity rules**. ...
Hi, firstly, check and if not present create a forwarding email, work.backup@company.com with status as pending. Then search for the threads and tell me full thread details, especially how many emails are under the thread related to the emergency contact. Modify another thread related to new business partnerships to la...
[ "get_thread", "list_threads", "list_labels", "create_forwarding_address", "list_forwarding_addresses", "get_user_profile", "modify_thread" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/email/dbs/db_1764696098233_cvtva4yqg.sql", "context": {"x-email-user-token": "TmnuRCr2rqp9at-Rf86fAz3q1sQrePp6-0TtCFBDV_A", "user_id": "bec5bb2c-807e-4f69-8358-9ddbe4ef9ca5", "pri...
[{"verifier_type": "database_state", "name": "Verify forwarding email added", "description": "Verify forwarding email added", "gym_name": "gym-email-mcp", "validation_config": {"query": "SELECT COUNT(*) AS cnt FROM forwarding_addresses fa WHERE fa.user_id = 'bec5bb2c-807e-4f69-8358-9ddbe4ef9ca5' AND fa.forwarding_email...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"email","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":3,"verification_results":{"Verify forwarding email added":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) AS cnt FROM for...
null
Hi, firstly, check and if not present create a forwarding email, work.backup@company.com with status as pending. Then search for the threads and tell me full thread details, especially how many emails are under the thread related to the emergency contact. Modify another thread related to new business partnerships to la...
8
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-email_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260108_050214_962_e5bf1251_0cabbd3d.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
69,010
3,302
4,201
swes-gpt-5.5-2026-04-23-oh-getmoto__moto.694ce1f4.pr_5093
diff --git a/moto/mediaconnect/models.py b/moto/mediaconnect/models.py index bbb34e7..7b46be4 100644 --- a/moto/mediaconnect/models.py +++ b/moto/mediaconnect/models.py @@ -123,6 +123,12 @@ class MediaConnectBackend(BaseBackend): sources: List[Dict[str, Any]], vpc_interfaces: List[Dict[str, Any]], ...
[ "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_alternative_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_list_flows_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_describe_flow_succeeds...
[ "tests/test_mediaconnect/test_mediaconnect.py::test_unknown_flow", "tests/test_mediaconnect/test_mediaconnect.py::test_tag_resource_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_add_flow_vpc_interfaces_fails", "tests/test_mediaconnect/test_mediaconnect.py::test_remove_flow_vpc_interface_fails...
swebench/swesmith.x86_64.getmoto_1776_moto.694ce1f4
swesmith/getmoto__moto.694ce1f4
Flow Creation and Update Issues in MediaConnect Description After the recent changes, there seems to be an issue with creating and updating flows in the MediaConnect service. When attempting to create a flow, the expected outputs and source configurations are not being set correctly. Specifically, the `ListenerAddres...
getmoto__moto.694ce1f4.pr_5093
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/getmoto__moto.694ce1f4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Flow Creation and Update Issues in MediaConnect Description After the recent changes, there seems to be an...
56
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/getmoto__moto.694ce1f4.pr_5093.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/moto/mediaconnect/models.py b/moto/mediaconnect/models.py index 7b46be4..93ece10 100644 --- a/moto/mediaconnect/models.py +++ b/moto/mediaconnect/models.py @@ -82,7 +82,7 @@ class MediaConnectBackend(BaseBackend): f":{flow_id}:{source['name']}" ) if not source.get...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_succeeds", "tests/test_mediaconnect/test_mediaconnect.py::test_create_flow_alternative_succeeds", ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,177,364
5,004
17,811
swes-gemini-gemini-3.5-flash-oh-sqlfluff__sqlfluff.50a1c4b6.func_pm_remove_wrapper__d5vnt3l5
diff --git a/src/sqlfluff/cli/commands.py b/src/sqlfluff/cli/commands.py index 14e3c3e..71943bb 100644 --- a/src/sqlfluff/cli/commands.py +++ b/src/sqlfluff/cli/commands.py @@ -642,23 +642,6 @@ def lint( if verbose >= 1 and not non_human_output: click.echo(format_linting_result_header()) - with PathA...
[ "test/cli/commands_test.py::test__cli__command_directed", "test/cli/commands_test.py::test__cli__command_no_dialect[command2]", "test/cli/commands_test.py::test__cli__command_lint_stdin[command0]", "test/cli/commands_test.py::test__cli__command_lint_stdin[command1]", "test/cli/commands_test.py::test__cli__c...
[ "test/cli/commands_test.py::test__cli__command_dialect", "test/cli/commands_test.py::test__cli__command_no_dialect[command0]", "test/cli/commands_test.py::test__cli__command_no_dialect[command1]", "test/cli/commands_test.py::test__cli__command_no_dialect[command3]", "test/cli/commands_test.py::test__cli__co...
swebench/swesmith.x86_64.sqlfluff_1776_sqlfluff.50a1c4b6
swesmith/sqlfluff__sqlfluff.50a1c4b6
CLI lint command fails with NameError: name 'result' is not defined #### Description When running the `sqlfluff lint` command, it crashes with a `NameError` because the variable `result` is referenced but never defined in the current scope. #### Steps/Code to Reproduce ```bash sqlfluff lint test/fixtures/cli/passin...
sqlfluff__sqlfluff.50a1c4b6.func_pm_remove_wrapper__d5vnt3l5
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/sqlfluff__sqlfluff.50a1c4b6/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> CLI lint command fails with NameError: name 'result' is not defined #### Description When running the...
63
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/sqlfluff__sqlfluff.50a1c4b6.func_pm_remove_wrapper__d5vnt3l5.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/AGENTS.md b/AGENTS.md new file mode 100644 index 0000000..f0bca61 --- /dev/null +++ b/AGENTS.md @@ -0,0 +1,11 @@ +# SQLFluff CLI Commands Context + +## CLI Architecture +- The CLI command entrypoints are defined using **click** in `src/sqlfluff/cli/commands.py`. +- The `lint` command processes path argumen...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/cli/commands_test.py::test__cli__command_directed", "test/cli/commands_test.py::test__cli__command_no_dialect[command2]", "test/cli/commands_te...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,214,210
11,218
21,343
swes-gemini-gemini-3.5-flash-oh-bottlepy__bottle.a8dfef30.func_pm_class_rm_funcs__weo8gf3c
diff --git a/bottle.py b/bottle.py index 286b5bc..52ef6c4 100755 --- a/bottle.py +++ b/bottle.py @@ -1146,13 +1146,6 @@ class BaseRequest(object): #: Maximum size of memory buffer for :attr:`body` in bytes. MEMFILE_MAX = 102400 - def __init__(self, environ=None): - """ Wrap a WSGI environ dictiona...
[ "test/test_auth.py::TestBasicAuth::test__header", "test/test_contextlocals.py::TestThreadLocals::test_request", "test/test_environ.py::TestRequest::test_app_property", "test/test_environ.py::TestRequest::test_auth", "test/test_environ.py::TestRequest::test_bigbody", "test/test_environ.py::TestRequest::tes...
[ "test/test_app.py::TestApplicationObject::test_setattr", "test/test_config.py::TestConfDict::test_gc_overlays", "test/test_config.py::TestConfDict::test_isadict", "test/test_config.py::TestConfDict::test_load_dict", "test/test_config.py::TestConfDict::test_load_module", "test/test_config.py::TestConfDict:...
swebench/swesmith.x86_64.bottlepy_1776_bottle.a8dfef30
swesmith/bottlepy__bottle.a8dfef30
BaseRequest missing __init__ method causes TypeError on instantiation #### Description After recent changes, creating a BaseRequest instance fails with a TypeError because the `__init__` method has been removed. #### Steps/Code to Reproduce ```python import bottle # This should work but now raises TypeError reques...
bottlepy__bottle.a8dfef30.func_pm_class_rm_funcs__weo8gf3c
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/bottlepy__bottle.a8dfef30/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> BaseRequest missing __init__ method causes TypeError on instantiation #### Description After recent cha...
66
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/bottlepy__bottle.a8dfef30.func_pm_class_rm_funcs__weo8gf3c.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/bottle.py b/bottle.py index 52ef6c4..8b3ca00 100755 --- a/bottle.py +++ b/bottle.py @@ -1146,6 +1146,13 @@ class BaseRequest(object): #: Maximum size of memory buffer for :attr:`body` in bytes. MEMFILE_MAX = 102400 + def __init__(self, environ=None): + """ Wrap a WSGI environ dictiona...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "test/test_auth.py::TestBasicAuth::test__header", "test/test_contextlocals.py::TestThreadLocals::test_request", "test/test_environ.py::TestRequest::t...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,685,823
11,368
21,821
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-lincolnloop__python-qrcode.456b01d4.combine_file__9kz9zzgb
diff --git a/qrcode/image/styles/colormasks.py b/qrcode/image/styles/colormasks.py index 43186b2..58a919d 100644 --- a/qrcode/image/styles/colormasks.py +++ b/qrcode/image/styles/colormasks.py @@ -25,7 +25,7 @@ class QRColorMask: paint_color = back_color def initialize(self, styledPilImage, image): - ...
[ "qrcode/tests/test_qrcode_pil.py::test_render_styled_Image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image_path", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_drawer[CircleModuleDrawer]", "qrcode...
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qr...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
QRColorMask initialization and background pixel retrieval broken #### Description When using styled QR code images with color masks, the color mask initialization is broken and background pixel retrieval returns incorrect colors. This affects all styled QR code generation with custom color masks and module drawers. ...
lincolnloop__python-qrcode.456b01d4.combine_file__9kz9zzgb
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> QRColorMask initialization and background pixel retrieval broken #### Description When using ...
62
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.combine_file__9kz9zzgb.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/qrcode/image/styles/colormasks.py b/qrcode/image/styles/colormasks.py index 58a919d..43186b2 100644 --- a/qrcode/image/styles/colormasks.py +++ b/qrcode/image/styles/colormasks.py @@ -25,7 +25,7 @@ class QRColorMask: paint_color = back_color def initialize(self, styledPilImage, image): - ...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_qrcode_pil.py::test_render_styled_Image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image", "qrcode/tests/...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,278,261
13,600
22,149
swes-gpt-5.5-2026-04-23-oh-mozilla__bleach.73871d76.combine_file__5vejqplu
diff --git a/bleach/_vendor/html5lib/_tokenizer.py b/bleach/_vendor/html5lib/_tokenizer.py index d884801..be3c6e9 100644 --- a/bleach/_vendor/html5lib/_tokenizer.py +++ b/bleach/_vendor/html5lib/_tokenizer.py @@ -856,16 +856,16 @@ class HTMLTokenizer(object): def beforeAttributeNameState(self): data = s...
[ "tests/test_clean.py::test_html_is_lowercased", "tests/test_clean.py::test_comments[<!--comment-->text-True-text]", "tests/test_clean.py::test_comments[<!--comment-->text-False-<!--comment-->text]", "tests/test_clean.py::test_comments[text<!--comment-->-False-text<!--comment-->]", "tests/test_clean.py::test...
[ "tests/test_callbacks.py::TestNofollowCallback::test_blank", "tests/test_callbacks.py::TestNofollowCallback::test_no_href", "tests/test_callbacks.py::TestNofollowCallback::test_basic", "tests/test_callbacks.py::TestNofollowCallback::test_mailto", "tests/test_callbacks.py::TestNofollowCallback::test_has_nofo...
swebench/swesmith.x86_64.mozilla_1776_bleach.73871d76
swesmith/mozilla__bleach.73871d76
HTML tokenizer produces empty output for many inputs #### Description The HTML tokenizer is returning empty strings for various HTML inputs that should produce valid output. This affects both cleaning and linkifying operations. #### Steps/Code to Reproduce ```python import bleach # Basic cleaning should preserve c...
mozilla__bleach.73871d76.combine_file__5vejqplu
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/mozilla__bleach.73871d76/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> HTML tokenizer produces empty output for many inputs #### Description The HTML tokenizer is returning em...
60
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/mozilla__bleach.73871d76.combine_file__5vejqplu.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/bleach/_vendor/html5lib/_tokenizer.py b/bleach/_vendor/html5lib/_tokenizer.py index be3c6e9..d884801 100644 --- a/bleach/_vendor/html5lib/_tokenizer.py +++ b/bleach/_vendor/html5lib/_tokenizer.py @@ -856,16 +856,16 @@ class HTMLTokenizer(object): def beforeAttributeNameState(self): data = s...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_clean.py::test_html_is_lowercased", "tests/test_clean.py::test_comments[<!--comment-->text-True-text]", "tests/test_clean.py::test_commen...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,924,427
3,341
15,492
swes-gpt-5.5-2026-04-23-oh-andialbrecht__sqlparse.e57923b3.func_basic__ztx58hlb
diff --git a/sqlparse/filters/output.py b/sqlparse/filters/output.py index 253537e..e4081ed 100644 --- a/sqlparse/filters/output.py +++ b/sqlparse/filters/output.py @@ -20,13 +20,13 @@ class OutputFilter: def process(self, stmt): self.count += 1 - if self.count > 1: - varname = '{f.var...
[ "tests/test_format.py::TestOutputFormat::test_python", "tests/test_format.py::TestOutputFormat::test_python_multiple_statements", "tests/test_format.py::TestOutputFormat::test_php" ]
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_valid_args", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::t...
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
OutputFilter generates incorrect variable names for first statement #### Description When using sqlparse's output formatting with `output_format='python'` or `output_format='php'`, the first SQL statement gets an incorrect variable name. Instead of the expected base variable name (like `sql` or `$sql`), it generates ...
andialbrecht__sqlparse.e57923b3.func_basic__ztx58hlb
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> OutputFilter generates incorrect variable names for first statement #### Description When using s...
48
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_basic__ztx58hlb.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/filters/output.py b/sqlparse/filters/output.py index e4081ed..253537e 100644 --- a/sqlparse/filters/output.py +++ b/sqlparse/filters/output.py @@ -20,13 +20,13 @@ class OutputFilter: def process(self, stmt): self.count += 1 - if self.count >= 1: - varname = '{f.va...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_format.py::TestOutputFormat::test_python", "tests/test_format.py::TestOutputFormat::test_python_multiple_statements", "tests/test_format....
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,145,713
3,673
13,275
eog-gpt-5.5-2026-04-23-oh-task_20251222_131624_825_8ce75c22_535e7ced
null
null
null
null
null
Emergency situation - our storage infrastructure case for that NetApp FAS system (the 2750 model specifically) came in through the web portal and somehow got triaged in pending state as low priority. But the customer just called - they're losing data in real-time. Multiple production databases are corrupted. This needs...
task_20251222_131624_825_8ce75c22_535e7ced
hybrid
You are a highly capable and intelligent AI assistant designed to help users by executing tasks. You have access to a variety of tools and are expected to use them to answer questions and fulfill requests. **Your Thought Process:** 1. **Analyze the Request:** Carefully analyze the user's latest message to understand t...
Emergency situation - our storage infrastructure case for that NetApp FAS system (the 2750 model specifically) came in through the web portal and somehow got triaged in pending state as low priority. But the customer just called - they're losing data in real-time. Multiple production databases are corrupted. This needs...
[ "create_virtual_event_webinar", "list_channels", "find_contacts", "send_channel_message", "find_products", "search_cases", "update_case", "publish_virtual_event_webinar", "find_user", "send_notification" ]
[]
/mcp
1
true
[{"mcp_server_name": "sn-csm-server", "mcp_server_url": "http://localhost:8001", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/hybrid/dbs/db_1765826170153_yh400aejr.sql", "context": {"x-user-email": "joanne.simpson@servicenow.com", "x-access-token": "tok_f4d363b73d4b_83e7a19d05bc_0195c10f"}, "user_info": ...
[{"verifier_type": "database_state", "name": "Case Escalated to Critical", "description": "Confirms NetApp case escalated from low to critical with impact reason", "gym_name": "sn-csm-server", "validation_config": {"query": "SELECT COUNT(*) FROM customer_case WHERE case_id = 8 AND priority = 'critical' AND escalation =...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"hybrid","passed_verifiers":20,"resolved":false,"score":0,"total_verifiers":27,"verification_results":{"Case Escalated to Critical":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) FROM customer_c...
null
Emergency situation - our storage infrastructure case for that NetApp FAS system (the 2750 model specifically) came in through the web portal and somehow got triaged in pending state as low priority. But the customer just called - they're losing data in real-time. Multiple production databases are corrupted. This needs...
11
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-hybrid_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251222_131624_825_8ce75c22_535e7ced.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
41,657
2,090
3,412
swes-gemini-gemini-3.5-flash-oh-lincolnloop__python-qrcode.456b01d4.func_pm_remove_assign__nj3vt1gf
diff --git a/qrcode/release.py b/qrcode/release.py index 208ac1e..b2a29ab 100644 --- a/qrcode/release.py +++ b/qrcode/release.py @@ -16,7 +16,6 @@ def update_manpage(data): return base_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__))) - filename = os.path.join(base_dir, "doc", "qr.1")...
[ "qrcode/tests/test_release.py::test_invalid_data", "qrcode/tests/test_release.py::test_no_change", "qrcode/tests/test_release.py::test_change" ]
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qr...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
# Undefined variable 'filename' in update_manpage function ## Description The `update_manpage` function in `qrcode/release.py` has a bug where the `filename` variable is being used before it's defined. This causes the function to fail with a `NameError` when it tries to open the file. The issue appears to be that th...
lincolnloop__python-qrcode.456b01d4.func_pm_remove_assign__nj3vt1gf
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> # Undefined variable 'filename' in update_manpage function ## Description The `update_manpage...
37
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.func_pm_remove_assign__nj3vt1gf.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/AGENTS.md b/AGENTS.md new file mode 100644 index 0000000..d013357 --- /dev/null +++ b/AGENTS.md @@ -0,0 +1,8 @@ +# python-qrcode repository info + +## Setup and Testing +- **Environment:** Use the conda environment `testbed` for executing pytest. + - Python path: `/opt/miniconda3/envs/testbed/bin/python` ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_release.py::test_invalid_data", "qrcode/tests/test_release.py::test_no_change", "qrcode/tests/test_release.py::test_change" ...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
731,202
5,522
13,459
eog-gemini-gemini-3.5-flash-oh-task_20251203_080912_583_df0ecdeb_a3470951
null
null
null
null
null
We’ve had a lot of scattered discussions around our incident follow-ups, and I (James Wilson) want to clean this up so the TechCorp Solutions Team can stay focused. Please create a new channel called Incident Follow-Up Hub under the TechCorp Solutions Team and add me, Alice Johnson, and Bob Smith as members. Once it’s ...
task_20251203_080912_583_df0ecdeb_a3470951
teams
# **Teams Assistant Policy** You are a **Microsoft Teams Management Assistant**. Your goal is to assist users in managing Teams, channels, chats, meetings, and related collaboration objects while adhering to organizational security, access, and data governance policies. You must always act on **confirmed context** a...
We’ve had a lot of scattered discussions around our incident follow-ups, and I (James Wilson) want to clean this up so the TechCorp Solutions Team can stay focused. Please create a new channel called Incident Follow-Up Hub under the TechCorp Solutions Team and add me, Alice Johnson, and Bob Smith as members. Once it’s ...
[ "add_channel_member", "create_chat", "list_teams", "send_channel_message", "archive_channel", "create_call", "send_chat_message", "update_channel", "list_users", "create_channel", "list_channels" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-teams-mcp", "mcp_server_url": "http://localhost:8002", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/teams/dbs/db_1764274139477_ve41d08gm.sql", "context": {"x-teams-access-token": "ya29.A0ARrdaM-super999unique888token777special666.SuperUserAllPermissionsTokenAccess_99-99"}, "user...
[{"verifier_type": "database_state", "name": "Incident Follow-Up Hub Channel Created", "description": "Verify that the Incident Follow-Up Hub channel is created and active under TechCorp Solutions Team.", "gym_name": "gym-teams-mcp", "validation_config": {"query": "SELECT COUNT(*) AS count FROM channels WHERE display_n...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"teams","passed_verifiers":9,"resolved":false,"score":0,"total_verifiers":10,"verification_results":{"Alice Added To Incident Follow-Up Hub":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) AS cou...
null
We’ve had a lot of scattered discussions around our incident follow-ups, and I (James Wilson) want to clean this up so the TechCorp Solutions Team can stay focused. Please create a new channel called Incident Follow-Up Hub under the TechCorp Solutions Team and add me, Alice Johnson, and Bob Smith as members. Once it’s ...
12
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-teams_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251203_080912_583_df0ecdeb_a3470951.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
256,645
6,773
8,256
swes-gemini-gemini-3.5-flash-oh-kurtmckee__feedparser.cad965a3.func_pm_ctrl_shuffle__b0jrbo2u
diff --git a/feedparser/namespaces/psc.py b/feedparser/namespaces/psc.py index ed3e602..e9f02c3 100644 --- a/feedparser/namespaces/psc.py +++ b/feedparser/namespaces/psc.py @@ -43,12 +43,11 @@ class Namespace: super().__init__() def _start_psc_chapters(self, attrs_d): - context = self._get_contex...
[ "tests/test_well_formed.py::test_strict_parser[info1283]", "tests/test_well_formed.py::test_loose_parser[info1283]", "tests/test_well_formed.py::test_strict_parser[info1285]", "tests/test_well_formed.py::test_loose_parser[info1284]", "tests/test_well_formed.py::test_loose_parser[info1287]", "tests/test_we...
[ "tests/test_temporary_behavior.py::test_issue_328_fallback_behavior", "tests/test_make_safe_absolute_uri.py::test_catch_value_error[http://bad]test/]", "tests/test_feedparser_dict.py::test_single_mapping_target_2", "tests/test_feedparser_dict.py::test_license", "tests/test_feedparser_dict.py::test_category"...
swebench/swesmith.x86_64.kurtmckee_1776_feedparser.cad965a3
swesmith/kurtmckee__feedparser.cad965a3
UnboundLocalError when parsing feeds with psc:chapters elements #### Description When parsing RSS feeds that contain `psc:chapters` elements, feedparser raises an `UnboundLocalError: local variable 'context' referenced before assignment` error. #### Steps/Code to Reproduce ```python import feedparser # Sample RSS ...
kurtmckee__feedparser.cad965a3.func_pm_ctrl_shuffle__b0jrbo2u
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/kurtmckee__feedparser.cad965a3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> UnboundLocalError when parsing feeds with psc:chapters elements #### Description When parsing RSS ...
32
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/kurtmckee__feedparser.cad965a3.func_pm_ctrl_shuffle__b0jrbo2u.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/feedparser/namespaces/psc.py b/feedparser/namespaces/psc.py index e9f02c3..1462313 100644 --- a/feedparser/namespaces/psc.py +++ b/feedparser/namespaces/psc.py @@ -43,11 +43,11 @@ class Namespace: super().__init__() def _start_psc_chapters(self, attrs_d): + context = self._get_contex...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_well_formed.py::test_strict_parser[info1283]", "tests/test_well_formed.py::test_loose_parser[info1283]", "tests/test_well_formed.py::test...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
806,567
4,956
11,584
swes-gpt-5.5-2026-04-23-oh-lincolnloop__python-qrcode.456b01d4.combine_module__kuk9xsgj
diff --git a/qrcode/image/styles/colormasks.py b/qrcode/image/styles/colormasks.py index 43186b2..58a919d 100644 --- a/qrcode/image/styles/colormasks.py +++ b/qrcode/image/styles/colormasks.py @@ -25,7 +25,7 @@ class QRColorMask: paint_color = back_color def initialize(self, styledPilImage, image): - ...
[ "qrcode/tests/test_qrcode_pil.py::test_render_styled_Image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image_path", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_drawer[CircleModuleDrawer]", "qrcode...
[ "qrcode/tests/test_example.py::test_example", "qrcode/tests/test_qrcode.py::test_basic", "qrcode/tests/test_qrcode.py::test_large", "qrcode/tests/test_qrcode.py::test_invalid_version", "qrcode/tests/test_qrcode.py::test_invalid_border", "qrcode/tests/test_qrcode.py::test_overflow", "qrcode/tests/test_qr...
swebench/swesmith.x86_64.lincolnloop_1776_python-qrcode.456b01d4
swesmith/lincolnloop__python-qrcode.456b01d4
QR code styling broken with color masks and module drawers #### Description When using styled QR code generation with color masks or custom module drawers, the image generation fails with AttributeError. This affects all styled QR code functionality including color masks, embedded images, and custom module drawers. ...
lincolnloop__python-qrcode.456b01d4.combine_module__kuk9xsgj
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/lincolnloop__python-qrcode.456b01d4/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> QR code styling broken with color masks and module drawers #### Description When using styled...
36
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/lincolnloop__python-qrcode.456b01d4.combine_module__kuk9xsgj.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/qrcode/image/styles/colormasks.py b/qrcode/image/styles/colormasks.py index 58a919d..afe1c35 100644 --- a/qrcode/image/styles/colormasks.py +++ b/qrcode/image/styles/colormasks.py @@ -25,7 +25,7 @@ class QRColorMask: paint_color = back_color def initialize(self, styledPilImage, image): - ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "qrcode/tests/test_qrcode_pil.py::test_render_styled_Image", "qrcode/tests/test_qrcode_pil.py::test_render_styled_with_embeded_image", "qrcode/tests...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
931,575
2,721
12,091
eog-gemini-gemini-3.5-flash-oh-task_20251126_145836_025_be6c9105_3b572b98
null
null
null
null
null
Emily Davis needs to coordinate external access to the 'Employee Handbook.docx' for the annual compliance audit. The pending request from auditor@external.com should be approved with viewing rights only. After approval, grant the actual reader permission to auditor@external.com. John Smith no longer requires his curren...
task_20251126_145836_025_be6c9105_3b572b98
drive
You are a helpful Drive assistant with access to all available tools. Operate in a safe and fully authorized environment—you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform exactly one lookup per entity type, verify that you are reusing ...
Emily Davis needs to coordinate external access to the 'Employee Handbook.docx' for the annual compliance audit. The pending request from auditor@external.com should be approved with viewing rights only. After approval, grant the actual reader permission to auditor@external.com. John Smith no longer requires his curren...
[ "create_comments", "resolve_accessproposals", "update_files", "delete_permission", "list_comments", "list_files", "list_permission", "list_accessproposals" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-google-drive-mcp", "mcp_server_url": "http://localhost:8009", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/drive/dbs/db_1763731281343_wbypacphk.sql", "context": {"x-auth-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIwMUpCUU0zSzhUR0haWDlSTlZRTTVZMlA3VyIsImVtYWlsIjoiam9...
[{"verifier_type": "database_state", "name": "John Access Removed", "description": "John Access Removed", "gym_name": "gym-google-drive-mcp", "validation_config": {"query": "SELECT COUNT(*) AS cnt\nFROM permissions p\nINNER JOIN users u ON u.id = p.user_id\nWHERE p.file_id = '01JBQM3K8TGHZX9RNVQM5Y2P8T'\n AND u.email ...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"drive","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":8,"verification_results":{"Access Proposal Approved for External Auditor":{"actual":0,"comparison_type":"equals","details":"Comparison equals: 0 vs 1","expected":1,"passed":false,"query":"WITH auditor_p...
null
Emily Davis needs to coordinate external access to the 'Employee Handbook.docx' for the annual compliance audit. The pending request from auditor@external.com should be approved with viewing rights only. After approval, grant the actual reader permission to auditor@external.com. John Smith no longer requires his curren...
27
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-drive_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251126_145836_025_be6c9105_3b572b98.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
639,000
16,303
18,296
eog-gemini-gemini-3.5-flash-oh-task_20260105_101355_392_e0ac0c6d_4d94d938
null
null
null
null
null
I need help setting up calendars for my team. First, find the user with email 'bob.smith@techcorp.com' in the system. Once you have that user, check if they already have a calendar named 'Q4 Sprint Planning' - if it exists, update its timezone to 'America/Los_Angeles' and description to 'Quarter 4 sprint planning and r...
task_20260105_101355_392_e0ac0c6d_4d94d938
calendar
You are a highly capable and intelligent AI assistant designed to help users by executing tasks. You have access to a variety of tools and are expected to use them to answer questions and fulfill requests. **Your Thought Process:** 1. **Analyze the Request:** Carefully analyze the user's latest message to understand ...
I need help setting up calendars for my team. First, find the user with email 'bob.smith@techcorp.com' in the system. Once you have that user, check if they already have a calendar named 'Q4 Sprint Planning' - if it exists, update its timezone to 'America/Los_Angeles' and description to 'Quarter 4 sprint planning and r...
[ "update_calendar", "create_calendar", "get_calendar_from_list", "add_calendar_to_list", "get_user_by_email", "update_calendar_in_list", "get_calendar_list" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1767381239880_ibbsoh6gc.sql", "context": {"x-hr-user-token": "ya29.A0ARrdaM-k9Vq7GzY2pL4mQf8sN1xT0bR3uHcJWv5yKzP6eF2.qwErTyUIopASDfGhJkLzXcVbNm12_34-56"}, "user_inf...
[{"verifier_type": "database_state", "name": "User Found by Email", "description": "Verifies that the user was found using the get_user_by_email tool with the correct email address, ensuring the model doesn't assume user_id values.", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) as count F...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":1,"resolved":false,"score":0,"total_verifiers":6,"verification_results":{"Code Review Schedule Calendar Exists":{"actual":0,"comparison_type":"greater_than","details":"Comparison greater_than: 0 vs 0","expected":0,"passed":false,"query":"SELECT C...
null
I need help setting up calendars for my team. First, find the user with email 'bob.smith@techcorp.com' in the system. Once you have that user, check if they already have a calendar named 'Q4 Sprint Planning' - if it exists, update its timezone to 'America/Los_Angeles' and description to 'Quarter 4 sprint planning and r...
28
runs/enterprise_ops_gym_valid/google/gemini-3.5-flash/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260105_101355_392_e0ac0c6d_4d94d938.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gemini/gemini-3.5-flash
null
null
416,141
12,388
14,185
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-dask__dask.5f61e423.pr_10125
diff --git a/dask/dataframe/utils.py b/dask/dataframe/utils.py index 7260ae0..60bf328 100644 --- a/dask/dataframe/utils.py +++ b/dask/dataframe/utils.py @@ -457,31 +457,13 @@ def _check_dask(dsk, check_names=True, check_dtypes=True, result=None, scheduler check_dtypes=check_dtypes, res...
[ "dask/dataframe/tests/test_dataframe.py::test_Dataframe", "dask/dataframe/tests/test_dataframe.py::test_head_npartitions", "dask/dataframe/tests/test_dataframe.py::test_Series", "dask/dataframe/tests/test_dataframe.py::test_rename_columns", "dask/dataframe/tests/test_dataframe.py::test_describe_numeric[dask...
[ "dask/dataframe/tests/test_dataframe.py::test_head_tail", "dask/dataframe/tests/test_dataframe.py::test_head_npartitions_warn", "dask/dataframe/tests/test_dataframe.py::test_index_head", "dask/dataframe/tests/test_dataframe.py::test_Index", "dask/dataframe/tests/test_dataframe.py::test_axes", "dask/datafr...
swebench/swesmith.x86_64.dask_1776_dask.5f61e423
swesmith/dask__dask.5f61e423
AttributeError: 'Scalar' object not found in dask.dataframe.core It seems there's an issue with the recent changes in the Dask DataFrame module. When attempting to perform operations that involve scalar values, an AttributeError is raised, indicating that the module 'dask.dataframe.core' has no attribute 'Scalar'. Thi...
dask__dask.5f61e423.pr_10125
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/dask__dask.5f61e423/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> AttributeError: 'Scalar' object not found in dask.dataframe.core It seems there's an issue with the recent cha...
70
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/dask__dask.5f61e423.pr_10125.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/dask/dataframe/utils.py b/dask/dataframe/utils.py index 60bf328..7ac517e 100644 --- a/dask/dataframe/utils.py +++ b/dask/dataframe/utils.py @@ -457,7 +457,22 @@ def _check_dask(dsk, check_names=True, check_dtypes=True, result=None, scheduler check_dtypes=check_dtypes, resu...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "timed_out": true, "timeout": 90 }
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,667,441
13,596
23,704
eog-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-task_20260108_101001_581_701c5774_3ac4e4fb
null
null
null
null
null
Please review my vacation responder settings to make sure they reflect my preferences. Enable my out-of-office auto-reply starting December 15, 2025, and have it automatically turn off on January 1 of the following year. Subject: “Out of office – new year holidays” Message: “Hello, I’m on vacation in Bali. Please do no...
task_20260108_101001_581_701c5774_3ac4e4fb
email
# **Email Management Assistant Policy** **Role:** Email Management Assistant **Mandate:** Secure, accurate, and efficient management of user email communications (Messages, Drafts, Threads, Labels, Filters, and Settings). You must act solely based on **confirmed user context** and **existing data integrity rules**. ...
Please review my vacation responder settings to make sure they reflect my preferences. Enable my out-of-office auto-reply starting December 15, 2025, and have it automatically turn off on January 1 of the following year. Subject: “Out of office – new year holidays” Message: “Hello, I’m on vacation in Bali. Please do no...
[ "update_vacation_settings", "create_label", "get_vacation_settings", "create_filter", "list_labels", "list_filters", "send_message" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/email/dbs/db_1764696098233_cvtva4yqg.sql", "context": {"x-email-user-token": "Kx5XwZK6imPxgyAE88USnS4lF3np3br81Tf9FZr4Vno", "user_id": "104daf4b-2023-49f1-8def-39b4e9dcda5e", "pri...
[{"verifier_type": "database_state", "name": "Check that the vacation settings are enabled and set correctly", "description": "Check that the vacation settings are enabled and set correctly", "gym_name": "gym-email-mcp", "validation_config": {"query": "select count(*) from vacation_settings where user_id = '104daf4b-20...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"email","passed_verifiers":0,"resolved":false,"score":0,"total_verifiers":4,"verification_results":{"Check that the main text is as requested":{"actual":[{"data":""},{"data":""},{"data":""}],"comparison_type":"equals","details":"Comparison equals: [{'data': ''}, {'data': ''},...
null
Please review my vacation responder settings to make sure they reflect my preferences. Enable my out-of-office auto-reply starting December 15, 2025, and have it automatically turn off on January 1 of the following year. Subject: “Out of office – new year holidays” Message: “Hello, I’m on vacation in Bali. Please do no...
15
runs/enterprise_ops_gym_valid/minimax/minimax-M3/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-email_validation/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260108_101001_581_701c5774_3ac4e4fb.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
fireworks_ai/accounts/fireworks/models/minimax-m3
null
null
94,646
7,848
10,416
eog-gpt-5.5-2026-04-23-oh-task_20251204_115043_305_e81f6083_4acfd9ea
null
null
null
null
null
We're merging two teams next month, and several of our calendars are outdated, so begin by checking all calendars I currently own and delete every calendar except my primary calendar and the existing team calendar. Archive the team calendar by updating its name and description to clearly indicate that it has been archi...
task_20251204_115043_305_e81f6083_4acfd9ea
calendar
You are a Google Calendar automation agent with full administrative permissions to manage users, meetings, recordings, and messages. Operate in a safe and fully authorized environment — you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform...
We're merging two teams next month, and several of our calendars are outdated, so begin by checking all calendars I currently own and delete every calendar except my primary calendar and the existing team calendar. Archive the team calendar by updating its name and description to clearly indicate that it has been archi...
[ "create_calendar", "update_calendar_in_list", "get_calendar_list", "list_acl_rules", "delete_acl_rule", "get_colors", "delete_calendar", "insert_acl_rule", "watch_acl", "list_settings", "patch_calendar", "watch_events" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-calendar", "mcp_server_url": "http://localhost:8003", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/calendar/dbs/db_1764361046464_47vxzhsli.sql", "context": {"x-access-token": "ya29.A0ARrdaM-k9Vq7GzY2pL4mQf8sN1xT0bR3uHcJWv5yKzP6eF2.qwErTyUIopASDfGhJkLzXcVbNm12_34-56"}, "user_info...
[{"verifier_type": "database_state", "name": "verify_new_calendar_bob_writer_access", "description": "Verify 'Tech Pro New' calendar has Bob with writer access (from alice-projects)", "gym_name": "gym-calendar", "validation_config": {"query": "SELECT COUNT(*) FROM acls a JOIN calendars c ON a.calendar_id = c.calendar_i...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"calendar","passed_verifiers":9,"resolved":false,"score":0,"total_verifiers":10,"verification_results":{"verify_acl_watch_on_new_calendar":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"SELECT COUNT(*) FROM wat...
null
We're merging two teams next month, and several of our calendars are outdated, so begin by checking all calendars I currently own and delete every calendar except my primary calendar and the existing team calendar. Archive the team calendar by updating its name and description to clearly indicate that it has been archi...
31
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-calendar_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20251204_115043_305_e81f6083_4acfd9ea.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
120,535
14,066
16,705
swes-gemini-gemini-3.5-flash-oh-andialbrecht__sqlparse.e57923b3.func_basic__73lbf87d
diff --git a/sqlparse/sql.py b/sqlparse/sql.py index 1037375..aab06db 100644 --- a/sqlparse/sql.py +++ b/sqlparse/sql.py @@ -232,24 +232,26 @@ class TokenList(Token): def _token_matching(self, funcs, start=0, end=None, reverse=False): """next token that match functions""" if start is None: - ...
[ "tests/test_cli.py::test_valid_args", "tests/test_format.py::TestFormat::test_strip_comments_single", "tests/test_format.py::TestFormat::test_strip_comments_multi", "tests/test_format.py::TestFormat::test_strip_comments_preserves_linebreak", "tests/test_format.py::TestFormat::test_strip_comments_preserves_w...
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::test_invalid_outfile", "tests/test_cli....
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
Token matching returns incorrect index when no match found When using token matching functions in sqlparse, the method returns -1 instead of None when no matching token is found. This breaks existing code that expects None as the "no match" indicator. #### Steps/Code to Reproduce ```python import sqlparse # Parse a...
andialbrecht__sqlparse.e57923b3.func_basic__73lbf87d
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Token matching returns incorrect index when no match found When using token matching functions in ...
61
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_basic__73lbf87d.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/sql.py b/sqlparse/sql.py index aab06db..db202fa 100644 --- a/sqlparse/sql.py +++ b/sqlparse/sql.py @@ -232,13 +232,13 @@ class TokenList(Token): def _token_matching(self, funcs, start=0, end=None, reverse=False): """next token that match functions""" if start is None: - ...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_cli.py::test_valid_args", "tests/test_format.py::TestFormat::test_strip_comments_single", "tests/test_format.py::TestFormat::test_strip_c...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
2,240,768
21,789
32,762
swes-gpt-5.5-2026-04-23-oh-andialbrecht__sqlparse.e57923b3.lm_rewrite__c06hay4e
diff --git a/sqlparse/sql.py b/sqlparse/sql.py index 1037375..3acd469 100644 --- a/sqlparse/sql.py +++ b/sqlparse/sql.py @@ -309,36 +309,38 @@ class TokenList(Token): start = start if isinstance(start, int) else self.token_index(start) return start + self.tokens[start:].index(token) - def group_t...
[ "tests/test_format.py::TestFormatReindent::test_identifier_list_with_wrap_after", "tests/test_format.py::TestFormatReindent::test_long_identifier_list_with_functions", "tests/test_grouping.py::test_grouping_identifier_list_other", "tests/test_grouping.py::test_grouping_identifier_list_with_inline_comments", ...
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_valid_args", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::t...
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
Grouping of tokens not working as expected When using the `group_tokens` method to group a list of tokens, the behavior is not consistent with previous versions. Specifically, when attempting to group tokens with inline comments or ordering, the expected grouping does not occur, leading to unexpected results. For exa...
andialbrecht__sqlparse.e57923b3.lm_rewrite__c06hay4e
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Grouping of tokens not working as expected When using the `group_tokens` method to group a list of...
47
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.lm_rewrite__c06hay4e.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/sql.py b/sqlparse/sql.py index 3acd469..1ec1b37 100644 --- a/sqlparse/sql.py +++ b/sqlparse/sql.py @@ -313,34 +313,31 @@ class TokenList(Token): """Replace tokens by an instance of *grp_cls*.""" if not isinstance(start, int): start = self.token_index(start) - + ...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_format.py::TestFormatReindent::test_identifier_list_with_wrap_after", "tests/test_format.py::TestFormatReindent::test_long_identifier_list_with_functions...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,594,474
4,485
14,343
swes-gemini-gemini-3.5-flash-oh-iterative__dvc.1d6ea681.pr_8405
diff --git a/dvc/stage/cache.py b/dvc/stage/cache.py index f97b471..67bc51a 100644 --- a/dvc/stage/cache.py +++ b/dvc/stage/cache.py @@ -187,9 +187,9 @@ class StageCache: assert os.path.exists(parent) assert os.path.isdir(parent) dump_yaml(tmp, cache) - self.repo.cache.legacy.move(tmp,...
[ "tests/unit/stage/test_cache.py::test_stage_cache", "tests/unit/stage/test_cache.py::test_stage_cache_params", "tests/unit/stage/test_cache.py::test_stage_cache_wdir", "tests/unit/stage/test_cache.py::test_shared_stage_cache", "tests/unit/stage/test_run.py::test_run_stage_dry[cmd1-expected1]" ]
[ "tests/unit/stage/test_cache.py::test_unhashable[kwargs0]", "tests/unit/stage/test_cache.py::test_unhashable[kwargs1]", "tests/unit/stage/test_cache.py::test_unhashable[kwargs2]", "tests/unit/stage/test_cache.py::test_unhashable[kwargs3]", "tests/unit/stage/test_cache.py::test_unhashable[kwargs4]" ]
swebench/swesmith.x86_64.iterative_1776_dvc.1d6ea681
swesmith/iterative__dvc.1d6ea681
# Stage cache restore function doesn't handle dry run parameter correctly #### Problem Description When using the stage cache restore functionality, there's an issue with the `dry` parameter handling. The `restore` method in `StageCache` class no longer accepts the `dry` parameter, but the `run_stage` function still ...
iterative__dvc.1d6ea681.pr_8405
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/iterative__dvc.1d6ea681/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> # Stage cache restore function doesn't handle dry run parameter correctly #### Problem Description When u...
99
runs/swesmith2k_valid64/google/gemini-3.5-flash/SWE-bench__SWE-smith-py-train/gemini/gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/iterative__dvc.1d6ea681.pr_8405.tar.gz
gemini-3.5-flash_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/dvc/stage/cache.py b/dvc/stage/cache.py index 67bc51a..f97b471 100644 --- a/dvc/stage/cache.py +++ b/dvc/stage/cache.py @@ -187,9 +187,9 @@ class StageCache: assert os.path.exists(parent) assert os.path.isdir(parent) dump_yaml(tmp, cache) - self.repo.odb.local.move(tmp, pa...
gemini/gemini-3.5-flash
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/unit/stage/test_cache.py::test_stage_cache", "tests/unit/stage/test_cache.py::test_stage_cache_params", "tests/unit/stage/test_cache.py::test_...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
5,094,135
34,815
47,450
swes-fireworks_ai-accounts-fireworks-models-minimax-m3-oh-getnikola__nikola.0f4c230e.func_pm_remove_assign__dza7o219
diff --git a/nikola/packages/pygments_better_html/__init__.py b/nikola/packages/pygments_better_html/__init__.py index ec5e62b..efb0abf 100644 --- a/nikola/packages/pygments_better_html/__init__.py +++ b/nikola/packages/pygments_better_html/__init__.py @@ -87,7 +87,6 @@ class BetterHtmlFormatter(HtmlFormatter): ...
[ "tests/integration/test_archive_full.py::test_avoid_double_slash_in_rss", "tests/integration/test_archive_full.py::test_check_links", "tests/integration/test_archive_full.py::test_index_in_sitemap", "tests/integration/test_archive_full.py::test_full_archive[overall]", "tests/integration/test_archive_full.py...
[ "tests/integration/test_archive_full.py::test_check_files", "tests/integration/test_archive_per_day.py::test_check_files", "tests/integration/test_archive_per_month.py::test_check_files", "tests/integration/test_building_in_subdir.py::test_check_files", "tests/integration/test_category_destpath.py::test_arc...
swebench/swesmith.x86_64.getnikola_1776_nikola.0f4c230e
swesmith/getnikola__nikola.0f4c230e
BetterHtmlFormatter line numbering broken after code cleanup #### Description The `BetterHtmlFormatter` in `nikola/packages/pygments_better_html/__init__.py` appears to have broken line numbering functionality. When using syntax highlighting with line numbers, the formatter fails to properly generate the HTML output....
getnikola__nikola.0f4c230e.func_pm_remove_assign__dza7o219
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/getnikola__nikola.0f4c230e/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> BetterHtmlFormatter line numbering broken after code cleanup #### Description The `BetterHtmlFormatter...
56
runs/swesmith2k_valid64/minimax/minimax-M3/SWE-bench__SWE-smith-py-train/fireworks_ai/accounts/fireworks/models/minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/getnikola__nikola.0f4c230e.func_pm_remove_assign__dza7o219.tar.gz
minimax-m3_sdk_513653c_maxiter_100_N_sdk-logprobs-all
false
diff --git a/nikola/packages/pygments_better_html/__init__.py b/nikola/packages/pygments_better_html/__init__.py index efb0abf..ec5e62b 100644 --- a/nikola/packages/pygments_better_html/__init__.py +++ b/nikola/packages/pygments_better_html/__init__.py @@ -87,6 +87,7 @@ class BetterHtmlFormatter(HtmlFormatter): ...
fireworks_ai/accounts/fireworks/models/minimax-m3
{ "patch_exists": true, "resolved": false, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/integration/test_archive_full.py::test_avoid_double_slash_in_rss", "tests/integration/test_archive_full.py::test_check_links", "tests/integra...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
1,329,048
8,849
21,389
eog-gpt-5.5-2026-04-23-oh-task_20260109_034708_645_be6c9105_8b22bcd8
null
null
null
null
null
Emily needs a structured workspace for the employee handbook revision project. The workspace must include a dedicated 'Project Phoenix' folder containing the existing 'Employee Handbook.docx' file, with the HR Department (hr@company.com) team having full editing capabilities. When moving the file to the folder, update ...
task_20260109_034708_645_be6c9105_8b22bcd8
drive
You are a helpful Drive assistant with access to all available tools. Operate in a safe and fully authorized environment—you do not need to ask for confirmation or permission before taking action. When identifiers such as names or IDs are missing, perform exactly one lookup per entity type, verify that you are reusing ...
Emily needs a structured workspace for the employee handbook revision project. The workspace must include a dedicated 'Project Phoenix' folder containing the existing 'Employee Handbook.docx' file, with the HR Department (hr@company.com) team having full editing capabilities. When moving the file to the folder, update ...
[ "list_comments", "create_permission", "list_files", "update_files", "create_comments", "create_file" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-google-drive-mcp", "mcp_server_url": "http://localhost:8009", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/drive/dbs/db_1763988076613_f5g41taxz.sql", "context": {"x-auth-token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIwMUpCUU0zSzhUR0haWDlSTlZRTTVZMlA3WiIsImVtYWlsIjoiZW1...
[{"verifier_type": "database_state", "name": "Verify Folder Created", "description": "Verify Folder Created", "gym_name": "gym-google-drive-mcp", "validation_config": {"query": "WITH new_folder AS (\n SELECT\n name,\n mime_type,\n trashed\n FROM\n files\n WHERE\n name = 'Proj...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"drive","passed_verifiers":5,"resolved":true,"score":1,"total_verifiers":5,"verification_results":{"Verify Comment Acknowledges Context":{"actual":1,"comparison_type":"equals","details":"Comparison equals: 1 vs 1","expected":1,"passed":true,"query":"WITH project_comment AS (\...
null
Emily needs a structured workspace for the employee handbook revision project. The workspace must include a dedicated 'Project Phoenix' folder containing the existing 'Employee Handbook.docx' file, with the HR Department (hr@company.com) team having full editing capabilities. When moving the file to the folder, update ...
8
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-drive_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260109_034708_645_be6c9105_8b22bcd8.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
true
null
gpt-5.5-2026-04-23
null
null
29,552
2,041
2,735
eog-gpt-5.5-2026-04-23-oh-task_20260108_101001_581_701c5774_3ac4e4fb
null
null
null
null
null
Please review my vacation responder settings to make sure they reflect my preferences. Enable my out-of-office auto-reply starting December 15, 2025, and have it automatically turn off on January 1 of the following year. Subject: “Out of office – new year holidays” Message: “Hello, I’m on vacation in Bali. Please do no...
task_20260108_101001_581_701c5774_3ac4e4fb
email
# **Email Management Assistant Policy** **Role:** Email Management Assistant **Mandate:** Secure, accurate, and efficient management of user email communications (Messages, Drafts, Threads, Labels, Filters, and Settings). You must act solely based on **confirmed user context** and **existing data integrity rules**. ...
Please review my vacation responder settings to make sure they reflect my preferences. Enable my out-of-office auto-reply starting December 15, 2025, and have it automatically turn off on January 1 of the following year. Subject: “Out of office – new year holidays” Message: “Hello, I’m on vacation in Bali. Please do no...
[ "update_vacation_settings", "create_label", "get_vacation_settings", "create_filter", "list_labels", "list_filters", "send_message" ]
[]
/mcp
1
true
[{"mcp_server_name": "gym-email-mcp", "mcp_server_url": "http://localhost:8004", "seed_database_file": "Domain Wise DBs and Task-DB Mappings/email/dbs/db_1764696098233_cvtva4yqg.sql", "context": {"x-email-user-token": "Kx5XwZK6imPxgyAE88USnS4lF3np3br81Tf9FZr4Vno", "user_id": "104daf4b-2023-49f1-8def-39b4e9dcda5e", "pri...
[{"verifier_type": "database_state", "name": "Check that the vacation settings are enabled and set correctly", "description": "Check that the vacation settings are enabled and set correctly", "gym_name": "gym-email-mcp", "validation_config": {"query": "select count(*) from vacation_settings where user_id = '104daf4b-20...
enterprise-ops-gym
{"database_cleanup_succeeded":true,"domain":"email","passed_verifiers":2,"resolved":false,"score":0,"total_verifiers":4,"verification_results":{"Check that the main text is as requested":{"actual":[],"comparison_type":"equals","details":"Comparison equals: [] vs Hi Steve, I’ve set up my vacation responder and listed yo...
null
Please review my vacation responder settings to make sure they reflect my preferences. Enable my out-of-office auto-reply starting December 15, 2025, and have it automatically turn off on January 1 of the following year. Subject: “Out of office – new year holidays” Message: “Hello, I’m on vacation in Bali. Please do no...
14
runs/enterprise_ops_gym_valid/openai/gpt-5.5-med/enterpriseops-gym-Brendan__EnterpriseOps-Gym-v2-email_validation/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval/conversations/task_20260108_101001_581_701c5774_3ac4e4fb.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-inline-eval
false
null
gpt-5.5-2026-04-23
null
null
186,367
9,707
11,830
swes-gpt-5.5-2026-04-23-oh-andialbrecht__sqlparse.e57923b3.func_basic__73lbf87d
diff --git a/sqlparse/sql.py b/sqlparse/sql.py index 1037375..aab06db 100644 --- a/sqlparse/sql.py +++ b/sqlparse/sql.py @@ -232,24 +232,26 @@ class TokenList(Token): def _token_matching(self, funcs, start=0, end=None, reverse=False): """next token that match functions""" if start is None: - ...
[ "tests/test_cli.py::test_valid_args", "tests/test_format.py::TestFormat::test_strip_comments_single", "tests/test_format.py::TestFormat::test_strip_comments_multi", "tests/test_format.py::TestFormat::test_strip_comments_preserves_linebreak", "tests/test_format.py::TestFormat::test_strip_comments_preserves_w...
[ "tests/test_cli.py::test_cli_main_empty", "tests/test_cli.py::test_parser_empty", "tests/test_cli.py::test_main_help", "tests/test_cli.py::test_invalid_choice", "tests/test_cli.py::test_invalid_args", "tests/test_cli.py::test_invalid_infile", "tests/test_cli.py::test_invalid_outfile", "tests/test_cli....
swebench/swesmith.x86_64.andialbrecht_1776_sqlparse.e57923b3
swesmith/andialbrecht__sqlparse.e57923b3
Token matching returns incorrect index when no match found When using token matching functions in sqlparse, the method returns -1 instead of None when no matching token is found. This breaks existing code that expects None as the "no match" indicator. #### Steps/Code to Reproduce ```python import sqlparse # Parse a...
andialbrecht__sqlparse.e57923b3.func_basic__73lbf87d
null
null
null
null
null
null
null
null
null
null
swe-smith
null
null
I have access to a code repository in the directory /workspace/andialbrecht__sqlparse.e57923b3/ . You can explore and modify files using the available tools. Consider the following issue description: <issue_description> Token matching returns incorrect index when no match found When using token matching functions in ...
38
runs/swesmith2k_valid64/openai/gpt-5.5/SWE-bench__SWE-smith-py-train/gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all/conversations/andialbrecht__sqlparse.e57923b3.func_basic__73lbf87d.tar.gz
gpt-5.5-2026-04-23_sdk_513653c_maxiter_100_N_sdk-logprobs-all
true
diff --git a/sqlparse/sql.py b/sqlparse/sql.py index aab06db..35ec002 100644 --- a/sqlparse/sql.py +++ b/sqlparse/sql.py @@ -238,7 +238,7 @@ class TokenList(Token): funcs = funcs[0] if reverse: - assert end is not None + assert end is None indexes = range(start...
gpt-5.5-2026-04-23
{ "patch_exists": true, "resolved": true, "tests_status": { "FAIL_TO_PASS": { "success": [ "tests/test_cli.py::test_valid_args", "tests/test_format.py::TestFormat::test_strip_comments_single", "tests/test_format.py::TestFormat::test_strip_c...
+ cd /testbed + : '>>>>> Start Test Output' + source /opt/miniconda3/bin/activate ++ _CONDA_ROOT=/opt/miniconda3 ++ . /opt/miniconda3/etc/profile.d/conda.sh +++ export CONDA_EXE=/opt/miniconda3/bin/conda +++ CONDA_EXE=/opt/miniconda3/bin/conda +++ export _CE_M= +++ _CE_M= +++ export _CE_CONDA= +++ _CE_CONDA= +++ export...
890,553
4,038
12,568