Datasets:
id stringlengths 13 13 | version stringclasses 1
value | split stringclasses 1
value | record_type stringclasses 3
values | primary_behavior stringclasses 9
values | secondary_behaviors listlengths 2 2 | language stringclasses 8
values | ecosystem stringclasses 8
values | task_type stringclasses 7
values | difficulty stringclasses 3
values | task dict | environment dict | behavioral_labels dict | provenance dict | quality dict | supervision unknown | fingerprint stringlengths 64 64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
jlab-v1-00001 | 1.0.0 | train | repair | test_verification | [
"bounded_code_editing",
"final_reporting"
] | php | composer/phpunit | security_fix | easy | {
"instruction": "Harden the session in the identity gateway against path traversal while preserving valid behavior. The affected entry point is SessionPolicy00000; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve the public API",
"Do not add dependencies",
... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/session/session_000_00.php",
"test_file": "tests/session/test_session_000_00.php",
"target_symbol": "SessionPolicy0... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"read_file",
"search",
"apply_patch",
"run_tests",
"run_linter",
... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"initial_attempt": {
"summary": "A bounded first attempt was applied after inspecting the target implementation.",
"result": "failed"
},
"failure_signal": {
"type": "targeted_test_failure",
"evidence": "The targeted test exposes a mismatch between the proposed behavior and the acceptance criteri... | 0cfa3684414418b08e41827bb161c782854cb87db7e6735f31a989331da10195 |
jlab-v1-00002 | 1.0.0 | train | trajectory | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | java | gradle/junit | bug_fix | medium | {
"instruction": "Fix the defect where the token drops the original error context in the identity gateway. The affected entry point is TokenPolicy00001; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update a targeted test",
... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/main/java/token/TokenService000.java",
"test_file": "src/test/java/token/TokenService000Test.java",
"target_symbol"... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Use a bounded, updateable plan proportional to task complexity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"... | 296bc08215c1989ec22c1ce9685308acd753dc02702d6b1f123ddf590b732236 |
jlab-v1-00003 | 1.0.0 | train | preference | test_verification | [
"bounded_code_editing",
"final_reporting"
] | php | composer/phpunit | refactoring | easy | {
"instruction": "Refactor the pagination in the identity gateway to remove duplicated validation logic without changing behavior. The affected entry point is PaginationPolicy00002; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Do not expose secrets in logs",
"Prese... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/pagination/pagination_000_02.php",
"test_file": "tests/pagination/test_pagination_000_02.php",
"target_symbol": "Pa... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior":... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"chosen_sequence": [
{
"tool": "search",
"arguments": {
"query": "PaginationPolicy00002|nullability_defect",
"path": "."
}
},
{
"tool": "read_file",
"arguments": {
"path": "src/pagination/pagination_000_02.php",
"start_line": 1,
"max_... | ec79f8cf768b86be677ac1cceecd31b3d2aa5d18f60e359151713b6f84bda1dc |
jlab-v1-00004 | 1.0.0 | train | trajectory | bounded_code_editing | [
"repository_grounding",
"test_verification"
] | php | composer/phpunit | code_review | medium | {
"instruction": "Review the proposed cache change in the identity gateway for correctness, scope, and regressions. The affected entry point is CachePolicy00003; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change database schema... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/cache/cache_000_03.php",
"test_file": "tests/cache/test_cache_000_03.php",
"target_symbol": "CachePolicy00003",
"... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"search",
"run_tests",
"git_diff"
],
"target_behavior": "Pr... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Produce the smallest architecture-aligned patch that satisfies the task.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 5468b57f3cca764cc1be120fa9fd809152bd57993f6c5f23783602c8d9c5cd20 |
jlab-v1-00005 | 1.0.0 | train | repair | repository_grounding | [
"tool_selection_execution",
"bounded_code_editing"
] | typescript | node/vitest | documentation_configuration | medium | {
"instruction": "Correct the configuration and operator guidance for idempotency keys in the identity gateway. The affected entry point is QueuePolicy00004; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing project abstraction... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/queue/queue_000_04.ts",
"test_file": "test/queue/test_queue_000_04.ts",
"target_symbol": "QueuePolicy00004",
"tes... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"read_file",
"update_plan",
"apply_patch",
"run_tests",
"git_diff"
... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"initial_attempt": {
"summary": "A bounded first attempt was applied after inspecting the target implementation.",
"result": "failed"
},
"failure_signal": {
"type": "incorrect_assumption",
"evidence": "Repository evidence contradicts the initial assumption about the data representation."
},
... | 462cffab89a5475109bcc46161bec4e459ba466267037debc9f17caa1b672d81 |
jlab-v1-00006 | 1.0.0 | train | trajectory | bounded_code_editing | [
"repository_grounding",
"test_verification"
] | php | composer/phpunit | refactoring | hard | {
"instruction": "Refactor the webhook in the identity gateway to remove duplicated error mapping without changing behavior. The affected entry point is WebhookPolicy00005; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve the public API",
"Do not add dependenc... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/webhook/webhook_000_05.php",
"test_file": "tests/webhook/test_webhook_000_05.php",
"target_symbol": "WebhookPolicy0... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"search",
"apply_patch",
"run_tests",
"run_tests",
"git_diff"
],
"targ... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Produce the smallest architecture-aligned patch that satisfies the task.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | a6f92c3930980e5cba4031c17b366520c068629f6aec1ab1c950087ed8eabff2 |
jlab-v1-00007 | 1.0.0 | train | trajectory | test_verification | [
"bounded_code_editing",
"final_reporting"
] | python | pytest/ruff | bug_fix | easy | {
"instruction": "Fix the defect where the permission misclassifies transient failures as permanent in the identity gateway. The affected entry point is PermissionPolicy00006; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/permission/permission_000_06.py",
"test_file": "tests/permission/test_permission_000_06.py",
"target_symbol": "Perm... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior":... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Verify the claimed behavior with targeted checks before completion.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 6d11a7808e16e2525b8fa58ce8bbfd3ab94f706466256ba35d48d525c2cc7d56 |
jlab-v1-00008 | 1.0.0 | train | trajectory | bounded_code_editing | [
"repository_grounding",
"test_verification"
] | python | pytest/ruff | bug_fix | medium | {
"instruction": "Fix the defect where the configuration accepts malformed input after normalization in the identity gateway. The affected entry point is ConfigurationPolicy00007; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Do not expose secrets in logs",
"Preserv... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/configuration/configuration_000_07.py",
"test_file": "tests/configuration/test_configuration_000_07.py",
"target_sy... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior": "Produce the smal... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Produce the smallest architecture-aligned patch that satisfies the task.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 3b151433c851cb5badda0e8c4916fed1f0b2545373566598323ac4444b0f4ad8 |
jlab-v1-00009 | 1.0.0 | train | preference | test_verification | [
"bounded_code_editing",
"final_reporting"
] | php | composer/phpunit | refactoring | easy | {
"instruction": "Refactor the parser in the identity gateway to remove duplicated retry decisions without changing behavior. The affected entry point is ParserPolicy00008; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change data... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/parser/parser_000_08.php",
"test_file": "tests/parser/test_parser_000_08.php",
"target_symbol": "ParserPolicy00008"... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
]... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"chosen_sequence": [
{
"tool": "list_directory",
"arguments": {
"path": "src/parser",
"depth": 2
}
},
{
"tool": "search",
"arguments": {
"query": "ParserPolicy00008|retry_safety_defect",
"path": "."
}
},
{
"tool": "read_fi... | 2ffe2ae431a41bc13d0c4478acfb0db017bd7f6ec3260d7f9d1d97bce1065882 |
jlab-v1-00010 | 1.0.0 | train | preference | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | typescript | node/vitest | bug_fix | easy | {
"instruction": "Fix the defect where the serializer mutates shared state across requests in the identity gateway. The affected entry point is SerializerPolicy00009; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing project ab... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/serializer/serializer_000_09.ts",
"test_file": "test/serializer/test_serializer_000_09.ts",
"target_symbol": "Seria... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior": "Choose the lowe... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"chosen_sequence": [
{
"tool": "search",
"arguments": {
"query": "SerializerPolicy00009|state_isolation_defect",
"path": "."
}
},
{
"tool": "read_file",
"arguments": {
"path": "src/serializer/serializer_000_09.ts",
"start_line": 1,
"m... | a653baf1d6ffb9385f7b553c05340b1f116d23fc17551cd471dffe9fdb66aea7 |
jlab-v1-00011 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | python | pytest/ruff | test_generation | medium | {
"instruction": "Add regression coverage for the case where the validator matches identifiers case-insensitively in the identity gateway. The affected entry point is ValidatorPolicy00010; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve the public API",
"Do n... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/validator/validator_000_10.py",
"test_file": "tests/validator/test_validator_000_10.py",
"target_symbol": "Validato... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"run_tests",
"git_diff"
],
"target_behavior":... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 331878e56ae8bedaae29ef2136d573abafb8c0e923127dcde37ecc13fd465b14 |
jlab-v1-00012 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | typescript | node/vitest | bug_fix | medium | {
"instruction": "Fix the defect where the retry marks partial work as successful in the identity gateway. The affected entry point is RetryPolicy00011; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update a targeted test",
... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/retry/retry_000_11.ts",
"test_file": "test/retry/test_retry_000_11.ts",
"target_symbol": "RetryPolicy00011",
"tes... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior": "Choose the lowe... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 16f768ca9f98de8f358e4894ea7068be25be0013c08a434f3e761994585f5108 |
jlab-v1-00013 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | python | pytest/ruff | bug_fix | medium | {
"instruction": "Fix the defect where the timeout uses a broad filesystem permission in the identity gateway. The affected entry point is TimeoutPolicy00012; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Do not expose secrets in logs",
"Preserve existing configurat... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/timeout/timeout_000_12.py",
"test_file": "tests/timeout/test_timeout_000_12.py",
"target_symbol": "TimeoutPolicy000... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behav... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "orient",
"observation": "A bounded directory view reduces path assumptions without reading the full repository.",
"action": {
"tool": "list_directory",
... | b2bdad6edcebd6d624f6d430d521847b72f5caa295243564bb70d7348e719431 |
jlab-v1-00014 | 1.0.0 | train | trajectory | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | php | composer/phpunit | refactoring | easy | {
"instruction": "Refactor the search in the identity gateway to remove duplicated authorization checks without changing behavior. The affected entry point is SearchPolicy00013; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/search/search_000_13.php",
"test_file": "tests/search/test_search_000_13.php",
"target_symbol": "SearchPolicy00013"... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Use a bounded, updateable plan proportional to task complexity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"... | 081c64cd5117f45ea13bcab3bb715435f3859c0fae1995ba3bd53ffb3ef32bd2 |
jlab-v1-00015 | 1.0.0 | train | trajectory | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | python | pytest/ruff | bug_fix | medium | {
"instruction": "Fix the defect where the index returns stale state after an update in the identity gateway. The affected entry point is IndexPolicy00014; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing project abstractions"... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/index/index_000_14.py",
"test_file": "tests/index/test_index_000_14.py",
"target_symbol": "IndexPolicy00014",
"te... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Use a bounded, updateable plan proportional to task complexity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"... | 0840dc8dfd7cb19f5d4eac635661950d70cf9761228c86c8c1d7edd0077d3b55 |
jlab-v1-00016 | 1.0.0 | train | trajectory | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | java | gradle/junit | refactoring | medium | {
"instruction": "Refactor the migration in the identity gateway to remove duplicated serialization logic without changing behavior. The affected entry point is MigrationPolicy00015; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve the public API",
"Do not add... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/main/java/migration/MigrationService000.java",
"test_file": "src/test/java/migration/MigrationService000Test.java",
... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"run_tests",
"git_diff"
],
... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Use a bounded, updateable plan proportional to task complexity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"... | 5e64bd0718b960fcf41686302e8c0012148f695ab0b612cc35d7f5ba4b4b9879 |
jlab-v1-00017 | 1.0.0 | train | preference | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | python | pytest/ruff | code_review | medium | {
"instruction": "Review the proposed upload change in the identity gateway for correctness, scope, and regressions. The affected entry point is UploadPolicy00016; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update a targeted ... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/upload/upload_000_16.py",
"test_file": "tests/upload/test_upload_000_16.py",
"target_symbol": "UploadPolicy00016",
... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"chosen_sequence": [
{
"tool": "list_directory",
"arguments": {
"path": "src/upload",
"depth": 2
}
},
{
"tool": "search",
"arguments": {
"query": "UploadPolicy00016|authorization_order_defect",
"path": "."
}
},
{
"tool": "... | 4a79d70459836cef6dc306cd0c77b9ef25b4c62354d2049223fc3dd060dcb81a |
jlab-v1-00018 | 1.0.0 | train | trajectory | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | python | pytest/ruff | bug_fix | medium | {
"instruction": "Fix the defect where the download silently truncates an oversized payload in the identity gateway. The affected entry point is DownloadPolicy00017; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Do not expose secrets in logs",
"Preserve existing con... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/download/download_000_17.py",
"test_file": "tests/download/test_download_000_17.py",
"target_symbol": "DownloadPoli... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Use a bounded, updateable plan proportional to task complexity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"... | f8da6d5009b2c7ed266b00085bcf32b0f57aefe1f1d05290354b7570d5e61f0c |
jlab-v1-00019 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | go | go test | bug_fix | easy | {
"instruction": "Fix the defect where the audit does not preserve the public error contract in the identity gateway. The affected entry point is AuditPolicy00018; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change database sche... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "internal/audit/audit_000_18.go",
"test_file": "internal/audit/test_audit_000_18.go",
"target_symbol": "AuditPolicy00018... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior": "Choose the lowe... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 4e22d595e74592eb4e696fa5bb49801e92d130a0a5656cc8c0e0d98adb42897c |
jlab-v1-00020 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | typescript | node/vitest | test_generation | medium | {
"instruction": "Add regression coverage for the case where the worker skips cleanup after cancellation in the identity gateway. The affected entry point is WorkerPolicy00019; keep the change aligned with the existing layered service boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing ... | {
"repository_id": "jl-synrepo-0000",
"repository_shape": "layered service",
"domain": "identity gateway",
"synthetic_commit": "ca3b1b9cb189a9a4fed58cd1d18e3e07d3e07b83",
"target_file": "src/worker/worker_000_19.ts",
"test_file": "test/worker/test_worker_000_19.ts",
"target_symbol": "WorkerPolicy00019",
... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior": "Choose the lowe... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 75783b1d97b3cbb913b150aa277836025f5eb4fbb792aef8e6a6ff5c342798ee |
jlab-v1-00021 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | python | pytest/ruff | feature_implementation | medium | {
"instruction": "Add guarded support for input size limits in the cache of the dataset pipeline. The affected entry point is CachePolicy00100; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update a targeted test",
"Avoid u... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/cache/cache_001_00.py",
"test_file": "tests/cache/test_cache_001_00.py",
"target_symbol": "CachePolicy00100",
"t... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"run_tests",
"git_diff"
]... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "orient",
"observation": "A bounded directory view reduces path assumptions without reading the full repository.",
"action": {
"tool": "list_directory",
... | 65a45e40e995ae0c6dcc0974a2a51d05150657b78cf6658225d928058e2339bb |
jlab-v1-00022 | 1.0.0 | train | trajectory | tool_selection_execution | [
"repository_grounding",
"test_verification"
] | go | go test | bug_fix | easy | {
"instruction": "Fix the defect where the queue matches identifiers case-insensitively in the dataset pipeline. The affected entry point is QueuePolicy00101; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Do not expose secrets in logs",
"Preserve existing configura... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "internal/queue/queue_001_01.go",
"test_file": "internal/queue/test_queue_001_01.go",
"target_symbol": "QueuePolicy0010... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior": "Choose the lowe... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Choose the lowest-cost valid tool and use schema-correct arguments.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 03b9df86130f7495db2bd71c8a859827d2c4d808b614f353891dc42ef9faa671 |
jlab-v1-00023 | 1.0.0 | train | trajectory | planning_decomposition | [
"repository_grounding",
"final_reporting"
] | php | composer/phpunit | code_review | medium | {
"instruction": "Review the proposed webhook change in the dataset pipeline for correctness, scope, and regressions. The affected entry point is WebhookPolicy00102; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change database s... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/webhook/webhook_001_02.php",
"test_file": "tests/webhook/test_webhook_001_02.php",
"target_symbol": "WebhookPolicy... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"update_plan",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Use a bounded, updateable plan proportional to task complexity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"... | 10ab418503a695b0172428e80c9cdd0f22ae808b05d909ef1437f686960e2b8b |
jlab-v1-00024 | 1.0.0 | train | preference | permission_scope_safety | [
"bounded_code_editing",
"final_reporting"
] | rust | cargo | bug_fix | hard | {
"instruction": "Fix the defect where the permission uses a broad filesystem permission in the dataset pipeline. The affected entry point is PermissionPolicy00103; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing project abs... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/permission/permission_001_03.rs",
"test_file": "tests/permission/test_permission_001_03.rs",
"target_symbol": "Per... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": true,
"expected_tool_sequence": [
"search",
"read_file",
"request_approval",
"search"
],
"target_behavior": "Respect scope and request approval only for mate... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"chosen_sequence": [
{
"tool": "search",
"arguments": {
"query": "PermissionPolicy00103|least_privilege_defect",
"path": "."
}
},
{
"tool": "read_file",
"arguments": {
"path": "src/permission/permission_001_03.rs",
"start_line": 1,
"m... | 30a37101dd224f7b70dbe2d5c38201db7fc8cf9aa216035d1de80ac526368d41 |
jlab-v1-00025 | 1.0.0 | train | trajectory | code_review_security | [
"repository_grounding",
"permission_scope_safety"
] | java | gradle/junit | test_generation | easy | {
"instruction": "Add regression coverage for the case where the configuration does not verify the persisted state in the dataset pipeline. The affected entry point is ConfigurationPolicy00104; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Preserve the public API",
... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/main/java/configuration/ConfigurationService001.java",
"test_file": "src/test/java/configuration/ConfigurationServic... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"read_file",
"review_diff",
"git_diff"
],
"target_behavior": "Report evid... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Report evidence-backed correctness or security findings with calibrated severity.",
"steps": [
{
"step": 1,
"phase": "orient",
"observation": "A bounded directory view reduces path assumptions without reading the full repository.",
"action": {
"tool": "list_d... | cc260a8e96b92f854c3516de7d7c7db0b4c05de65302c48d9c22cf443444527f |
jlab-v1-00026 | 1.0.0 | train | trajectory | test_verification | [
"bounded_code_editing",
"final_reporting"
] | java | gradle/junit | feature_implementation | easy | {
"instruction": "Add guarded support for cursor pagination in the parser of the dataset pipeline. The affected entry point is ParserPolicy00105; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update a targeted test",
"Avoid... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/main/java/parser/ParserService001.java",
"test_file": "src/test/java/parser/ParserService001Test.java",
"target_sy... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"run_tests",
"git_diff"
],
"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Verify the claimed behavior with targeted checks before completion.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | edf328f2fa8f0a977c20092b6f8c992cbda83fdc6c7e8bf2f5a0801ff0af7f32 |
jlab-v1-00027 | 1.0.0 | train | trajectory | repository_grounding | [
"tool_selection_execution",
"bounded_code_editing"
] | go | go test | feature_implementation | easy | {
"instruction": "Add guarded support for request cancellation in the serializer of the dataset pipeline. The affected entry point is SerializerPolicy00106; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Do not expose secrets in logs",
"Preserve existing configurati... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "internal/serializer/serializer_001_06.go",
"test_file": "internal/serializer/test_serializer_001_06.go",
"target_symbo... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"update_plan",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavior"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Ground every decision in repository evidence before editing.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
"arg... | f0571dd67fd03133239a4bd187fe9739de10147e6c0bf68b7732ce5b793cc128 |
jlab-v1-00028 | 1.0.0 | train | trajectory | failure_diagnosis_repair | [
"test_verification",
"tool_selection_execution"
] | php | composer/phpunit | refactoring | medium | {
"instruction": "Refactor the validator in the dataset pipeline to remove duplicated error mapping without changing behavior. The affected entry point is ValidatorPolicy00107; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/validator/validator_001_07.php",
"test_file": "tests/validator/test_validator_001_07.php",
"target_symbol": "Valid... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"diagnose_failure",
"apply_patch",
"run_tests"... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Classify failure evidence, revise the hypothesis, and repair without looping.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "sear... | d7ce0d6df8a034db03d68bbd5b1ff702356f644b929992c7f4c9a05187b7e00c |
jlab-v1-00029 | 1.0.0 | train | trajectory | bounded_code_editing | [
"repository_grounding",
"test_verification"
] | java | gradle/junit | bug_fix | easy | {
"instruction": "Fix the defect where the retry silently truncates an oversized payload in the dataset pipeline. The affected entry point is RetryPolicy00108; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing project abstract... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/main/java/retry/RetryService001.java",
"test_file": "src/test/java/retry/RetryService001Test.java",
"target_symbol... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"git_diff"
],
"target_behavi... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Produce the smallest architecture-aligned patch that satisfies the task.",
"steps": [
{
"step": 1,
"phase": "orient",
"observation": "A bounded directory view reduces path assumptions without reading the full repository.",
"action": {
"tool": "list_directory"... | 0453600a90a4c25fcba5be3fa12cde5197a07f39afe62b55ae1c0b0c1d6ff647 |
jlab-v1-00030 | 1.0.0 | train | preference | test_verification | [
"bounded_code_editing",
"final_reporting"
] | python | pytest/ruff | refactoring | easy | {
"instruction": "Refactor the timeout in the dataset pipeline to remove duplicated error mapping without changing behavior. The affected entry point is TimeoutPolicy00109; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Preserve the public API",
"Do not add dependen... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/timeout/timeout_001_09.py",
"test_file": "tests/timeout/test_timeout_001_09.py",
"target_symbol": "TimeoutPolicy00... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behavior":... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"chosen_sequence": [
{
"tool": "search",
"arguments": {
"query": "TimeoutPolicy00109|compatibility_defect",
"path": "."
}
},
{
"tool": "read_file",
"arguments": {
"path": "src/timeout/timeout_001_09.py",
"start_line": 1,
"max_lines": ... | a22b81e7a97a3b1496090646db0c55f8130d7dfb0986f468936c6fbe6535171e |
jlab-v1-00031 | 1.0.0 | train | trajectory | test_verification | [
"bounded_code_editing",
"final_reporting"
] | php | composer/phpunit | refactoring | hard | {
"instruction": "Refactor the search in the dataset pipeline to remove duplicated retry decisions without changing behavior. The affected entry point is SearchPolicy00110; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Keep backward compatibility",
"Add or update a... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/search/search_001_10.php",
"test_file": "tests/search/test_search_001_10.php",
"target_symbol": "SearchPolicy00110... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"search",
"apply_patch",
"run_tests",
"run_linter",
"run_tests",
"git_... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Verify the claimed behavior with targeted checks before completion.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "search",
... | 10e3390e5ac0ff8ce602baf42c4ba521d28a473f52e9e3035f96f6349f80f610 |
jlab-v1-00032 | 1.0.0 | train | trajectory | code_review_security | [
"repository_grounding",
"permission_scope_safety"
] | rust | cargo | code_review | hard | {
"instruction": "Review the proposed index change in the dataset pipeline for correctness, scope, and regressions. The affected entry point is IndexPolicy00111; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Do not expose secrets in logs",
"Preserve existing config... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/index/index_001_11.rs",
"test_file": "tests/index/test_index_001_11.rs",
"target_symbol": "IndexPolicy00111",
"t... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"search",
"review_diff",
"git_diff"
],
"target_behavior": "Report evidence-bac... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Report evidence-backed correctness or security findings with calibrated severity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "... | 1d492c5f6e548928f658eca918ab0707a24eab3854ff9df86ce352e1e68fd438 |
jlab-v1-00033 | 1.0.0 | train | trajectory | final_reporting | [
"test_verification",
"repository_grounding"
] | java | gradle/junit | bug_fix | medium | {
"instruction": "Fix the defect where the migration drops the original error context in the dataset pipeline. The affected entry point is MigrationPolicy00112; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Keep the patch reviewable",
"Do not change database schema... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/main/java/migration/MigrationService001.java",
"test_file": "src/test/java/migration/MigrationService001Test.java",
... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"list_directory",
"search",
"read_file",
"apply_patch",
"run_tests",
"run_linter",
"git_diff"
],
"target_behav... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Report changed behavior, verification evidence, and remaining uncertainty precisely.",
"steps": [
{
"step": 1,
"phase": "orient",
"observation": "A bounded directory view reduces path assumptions without reading the full repository.",
"action": {
"tool": "lis... | 3a7bda1c7e10086c57866118c982dd123ec7b26c7293163079c5f9061b242eda |
jlab-v1-00034 | 1.0.0 | train | trajectory | code_review_security | [
"repository_grounding",
"permission_scope_safety"
] | php | composer/phpunit | security_fix | easy | {
"instruction": "Harden the upload in the dataset pipeline against replay attacks while preserving valid behavior. The affected entry point is UploadPolicy00113; keep the change aligned with the existing modular monolith boundary.",
"constraints": [
"Preserve error semantics",
"Reuse existing project abstr... | {
"repository_id": "jl-synrepo-0001",
"repository_shape": "modular monolith",
"domain": "dataset pipeline",
"synthetic_commit": "ef7b2fe767f63190f6a3dad9a7c670937068df3f",
"target_file": "src/upload/upload_001_13.php",
"test_file": "tests/upload/test_upload_001_13.php",
"target_symbol": "UploadPolicy00113... | {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"should_ask_clarification": false,
"requires_approval": false,
"expected_tool_sequence": [
"search",
"read_file",
"read_file",
"review_diff",
"search",
"git_diff"
],
"target_behavior": "Report evidence-bac... | {
"origin": "synthetic",
"generation_method": "deterministic_scenario_composition",
"repository_materialized": false,
"execution_verified": false,
"tool_results": "expected_or_simulated",
"human_reviewed": false,
"release_stage": "research_preview"
} | {
"schema_validated": true,
"exact_duplicate": false,
"split_grouped_by_repository_family": true,
"execution_verified": false,
"quality_tier": "silver_structural"
} | {
"policy_summary": "Report evidence-backed correctness or security findings with calibrated severity.",
"steps": [
{
"step": 1,
"phase": "ground",
"observation": "The task names behavior but repository evidence is required before selecting an edit target.",
"action": {
"tool": "... | f8c1efb29590958c961168501522824747ce2ed0c32925c601920b2a6f5b7f6e |
- Why this dataset exists
- Release status and evidence level
- Intended behavioral effects
- Dataset composition
- Language and code coverage
- Record model
- Tool vocabulary
- Loading the dataset
- Recommended training use
- Evaluation protocol
- Quality assurance
- Data generation methodology
- Biases and limitations
- Misuse and out-of-scope use
- Roadmap to a Gold release
- Project files
- Versioning
- License
- Citation
- References
- About JumpLander
JL-AgentBehavior-10K
JL-AgentBehavior-10K is a 10,000-record, English-language research-preview dataset for studying and training the behavioral policy of repository-level coding agents.
The dataset does not treat a coding agent as a chatbot that maps a request directly to a block of code. It represents an agent as a policy operating across a sequence of observable decisions:
task
-> repository evidence
-> bounded plan
-> tool selection
-> scoped edit strategy
-> verification
-> failure diagnosis and repair
-> evidence-based final report
Each record is labeled by the behavior it is intended to influence, the type of supervision it provides, its provenance, and its verification level. The release includes trajectory supervision, chosen/rejected preference pairs, and failure-repair examples across eight programming-language ecosystems.
Version 1.0.0 is a deterministically generated synthetic policy corpus. Its schema, counts, identifiers, fingerprints, and repository-family split isolation are validated. The repositories are not materialized and the candidate patches or expected tool observations are not runtime-executed. Every record states
execution_verified: falseandquality_tier: silver_structural. Do not describe this release as a runtime-verified or human-reviewed Gold dataset.
Why this dataset exists
Many code instruction datasets supervise only the final answer. That is useful for code completion and isolated problem solving, but it leaves important agent behaviors underspecified:
- whether the model searches before editing;
- whether it grounds file and symbol references in repository evidence;
- whether the plan is proportional to the task;
- whether the selected tool and arguments are valid;
- whether the patch remains inside the requested boundary;
- whether tests are executed before a success claim;
- whether a failed attempt changes the hypothesis;
- whether an operation requires user approval;
- whether the final report distinguishes evidence from assumption.
JL-AgentBehavior-10K makes those behaviors explicit and measurable. It is intended as a controlled starting point for supervised fine-tuning, preference optimization, ablation studies, data-pipeline development, and agent-policy research.
It is not intended to replace repository execution, real issue/patch trajectories, hidden tests, human code review, or an independent evaluation benchmark.
Release status and evidence level
| Property | v1.0.0 status |
|---|---|
| Total records | 10,000 |
| Data origin | Deterministic synthetic scenario composition |
| Schema validation | Completed |
| Exact ID/fingerprint uniqueness | Completed |
| Split isolation | Grouped by synthetic repository family |
| Repository materialization | Not performed |
| Patch execution | Not performed |
| Test execution | Not performed |
| Human review | Not performed |
| Quality tier | silver_structural |
| Recommended role | Policy warm-start, controlled research, pipeline prototyping |
| Not recommended as | Runtime-verified correctness corpus or standalone benchmark |
The evidence level is encoded per record rather than implied by the dataset title. This is intentional: users should be able to filter by provenance when future verified releases are merged with this corpus.
Intended behavioral effects
The dataset is organized around nine primary behavioral axes. Each axis is associated with a measurable hypothesis rather than a general claim that the agent becomes βbetter.β
| Primary behavior | Records | Intended behavioral shift | Suggested evaluation metric |
|---|---|---|---|
repository_grounding |
1,500 | Search and inspect evidence before selecting an edit target | wrong-file edit rate; unsupported file-reference rate |
planning_decomposition |
1,200 | Use short, updateable plans proportional to task complexity | plan adherence; unnecessary step rate; time to first valid action |
tool_selection_execution |
1,500 | Select a valid, low-cost tool with schema-correct arguments | correct tool@1; invalid argument rate; redundant tool-call rate |
bounded_code_editing |
2,000 | Keep the proposed change aligned with the existing architecture and requested scope | scope violation rate; unnecessary changed files; patch applicability |
test_verification |
1,200 | Verify before making a success claim | false-success rate; targeted-test selection accuracy |
failure_diagnosis_repair |
1,200 | Use new failure evidence to revise the hypothesis instead of repeating an action | recovery success; repeated-action rate; attempts to recovery |
code_review_security |
800 | Produce evidence-backed review findings with calibrated severity | valid finding precision; unsupported finding rate |
permission_scope_safety |
400 | Request approval at real trust boundaries without over-refusing safe local work | unsafe action rate; unnecessary refusal/approval rate |
final_reporting |
200 | Report changed behavior, executed checks, and remaining uncertainty precisely | fabricated-verification rate; evidence completeness |
These effects are hypotheses. They must be tested against a base model and appropriate controls. The dataset itself does not establish causality.
Dataset composition
Splits
| Split | Records | Synthetic repository families |
|---|---|---|
| Train | 9,000 | 450 |
| Validation | 500 | 25 |
| Test | 500 | 25 |
| Total | 10,000 | 500 |
Records from one synthetic repository family appear in only one split. This prevents direct family leakage within the packaged splits. The test split is suitable for pipeline checks, but it should not be treated as a strong independent benchmark because it is generated by the same scenario system as the training data.
Supervision formats
| Record type | Records | Purpose |
|---|---|---|
trajectory |
7,000 | Structured observation/action/expected-observation sequences |
preference |
2,000 | Chosen versus rejected action sequences with an explicit behavior-level reason |
repair |
1,000 | Failed-attempt signals, updated diagnosis, bounded repair sequence, and success criteria |
Task distribution
| Task type | Records |
|---|---|
| Bug fixing | 3,000 |
| Feature implementation | 2,000 |
| Refactoring | 1,500 |
| Test generation | 1,200 |
| Code review | 1,000 |
| Security fixes | 800 |
| Documentation/configuration | 500 |
Difficulty distribution
| Difficulty | Records |
|---|---|
| Easy | 2,500 |
| Medium | 5,000 |
| Hard | 2,500 |
Difficulty is a generation label based on scenario composition. It is not calibrated against human completion time or model pass rate in v1.
Language and code coverage
All natural-language instructions and supervision annotations are in English.
| Code language | Records | Referenced ecosystem |
|---|---|---|
| Python | 2,500 | pytest / ruff |
| TypeScript | 2,500 | Node / Vitest-style commands |
| PHP | 1,500 | Composer / PHPUnit / PHPStan |
| Java | 1,500 | Gradle / JUnit |
| Go | 1,000 | go test / golangci-lint |
| Rust | 400 | Cargo / Clippy |
| C# | 300 | .NET / xUnit-style commands |
| C++ | 300 | CMake / CTest / clang-tidy |
The command strings are part of synthetic environment descriptions. They are not evidence that the referenced commands were executed.
Record model
Each row is a JSON object with five conceptual layers:
- Task contract β instruction, constraints, and acceptance criteria.
- Environment description β repository family, target path, test path, symbols, and candidate commands.
- Behavioral labels β intended policy, approval boundary, and expected tool order.
- Supervision β trajectory, preference pair, or repair trace.
- Evidence metadata β provenance, quality tier, execution status, and fingerprint.
Compact example
{
"id": "jlab-v1-00001",
"version": "1.0.0",
"split": "train",
"record_type": "trajectory",
"primary_behavior": "repository_grounding",
"secondary_behaviors": [
"tool_selection_execution",
"bounded_code_editing"
],
"language": "typescript",
"task_type": "bug_fix",
"difficulty": "medium",
"task": {
"instruction": "Fix the defect where the session returns stale state after an update in the identity gateway.",
"constraints": [
"Preserve the public API",
"Do not add dependencies",
"Limit changes to relevant files"
],
"acceptance_criteria": [
"The session no longer exhibits the specified behavior",
"The public contract remains compatible",
"A targeted verification path is identified",
"The final change remains inside the requested scope"
]
},
"environment": {
"repository_id": "jl-synrepo-0000",
"target_file": "src/session/session_000_00.ts",
"test_file": "test/session/test_session_000_00.ts",
"targeted_test_command": "npm test -- test/session/test_session_000_00.ts"
},
"behavioral_labels": {
"must_ground_before_edit": true,
"must_verify_before_success_claim": true,
"requires_approval": false,
"target_behavior": "Ground every decision in repository evidence before editing."
},
"provenance": {
"origin": "synthetic",
"repository_materialized": false,
"execution_verified": false,
"human_reviewed": false,
"release_stage": "research_preview"
},
"quality": {
"schema_validated": true,
"execution_verified": false,
"quality_tier": "silver_structural"
},
"supervision": {
"policy_summary": "Ground every decision in repository evidence before editing.",
"steps": [
{
"step": 1,
"phase": "ground",
"action": {
"tool": "search",
"arguments": {
"query": "SessionPolicy00000|cache_invalidation_defect",
"path": "."
}
}
}
]
},
"fingerprint": "<sha256>"
}
The actual rows contain the complete supervision object. See schema.json and docs/SCHEMA.md for field-level details.
Tool vocabulary
The release uses a small, implementation-neutral action vocabulary:
searchlist_directoryread_fileupdate_planapply_patchrun_testsrun_lintergit_diffdiagnose_failurereview_diffrequest_approvalcontinue_without_approval
These are abstract training actions, not a claim about a specific agent runtime. Before training, map them to the exact tool names, role boundaries, JSON schemas, and result formats used by the deployment harness. A mismatch between training-time and inference-time tool interfaces can reduce transfer.
Loading the dataset
Hugging Face Datasets
After publishing under the intended organization:
from datasets import load_dataset
dataset = load_dataset("jumplander/JL-AgentBehavior-10K")
print(dataset)
print(dataset["train"][0]["primary_behavior"])
Local JSONL
from datasets import load_dataset
dataset = load_dataset(
"json",
data_files={
"train": "data/train.jsonl",
"validation": "data/validation.jsonl",
"test": "data/test.jsonl",
},
)
Stream records
import json
with open("data/train.jsonl", encoding="utf-8") as handle:
for line in handle:
record = json.loads(line)
if record["primary_behavior"] == "failure_diagnosis_repair":
print(record["id"], record["record_type"])
Recommended training use
1. Start with a code-capable base model
This dataset is designed to influence control policy. It is too small and too synthetic to create repository-level coding ability from scratch. The base model should already understand code, common software-engineering concepts, and the relevant programming languages.
2. Preserve provenance during preprocessing
Do not discard these fields before deciding how the record is used:
record_typeprimary_behaviorprovenance.execution_verifiedprovenance.human_reviewedquality.quality_tier
Future Gold records should be upweighted relative to synthetic policy records only after evaluation confirms the weighting is useful.
3. Do not train on synthetic tool results as facts
Tool outputs and expected observations describe the intended policy structure. They are not executed evidence. A conservative training formatter should:
- mask the user task and environment from the loss;
- mask external tool results from the loss;
- apply loss to assistant decisions, tool calls, repair classifications, and final-report contracts;
- avoid teaching the model to fabricate a successful test result;
- preserve the distinction between
expected,simulated, andexecutedevidence.
4. Separate training objectives by record type
| Record type | Suitable objective |
|---|---|
trajectory |
Supervised fine-tuning on policy actions or next-action prediction |
preference |
DPO/IPO-style preference optimization or pairwise reranking |
repair |
Failure-conditioned SFT, next-action prediction, or repair-policy training |
The repository includes scripts/format_sft.py as a conservative reference converter. It is not a universal chat template; adapt it to the tokenizer and agent runtime being trained.
5. Mix with real execution data
For serious coding-agent training, combine this release with legally usable, deduplicated data containing:
- materialized repository snapshots;
- issue or task provenance;
- actual tool outputs;
- applicable patches;
- executed targeted and regression tests;
- review outcomes;
- explicit license records;
- contamination controls.
Use execution_verified and quality_tier to prevent synthetic and executed records from being silently mixed.
Evaluation protocol
Do not evaluate the model only on the packaged synthetic test split. Use unseen, independently constructed, executable repository tasks.
Core metrics
| Dimension | Metric examples |
|---|---|
| Task completion | resolved@1; hidden-test pass rate |
| Grounding | wrong-file edit rate; unsupported symbol reference rate |
| Scope | unrelated changed files; unnecessary diff size; API break rate |
| Tools | correct tool@1; invalid argument rate; redundant calls; tool cost |
| Verification | false-success rate; targeted-test choice; fabricated result rate |
| Repair | recovery success; repeated action rate; attempts to recovery |
| Review | valid finding precision; severity calibration; unsupported findings |
| Safety | unsafe action rate; correct approval request; unnecessary refusal |
| Reporting | verification completeness; uncertainty calibration |
| Efficiency | latency; generated tokens; tool calls; execution cost |
Minimum controlled comparison
Compare at least:
- the unchanged base model;
- the base model fine-tuned on a random 10K code-instruction control;
- the base model fine-tuned on the complete JL-AgentBehavior-10K mixture;
- behavior-specific ablations.
Recommended ablations:
full mixture
full minus repository_grounding
full minus planning_decomposition
full minus tool_selection_execution
full minus test_verification
full minus failure_diagnosis_repair
full minus permission_scope_safety
Run models in the same harness with the same tool schemas, budgets, decoding settings, and repository tasks. For paired binary outcomes, report confidence intervals and use a paired method such as bootstrap resampling or McNemarβs test where appropriate. Report adverse shifts, not only average task success.
Example falsifiable hypotheses
| ID | Hypothesis | Primary metric | Important adverse metric |
|---|---|---|---|
| H1 | Grounding supervision reduces wrong-file edits | wrong-file edit rate | extra search calls |
| H2 | Planning supervision improves multi-file task success | resolved@1 on multi-file tasks | latency and token cost |
| H3 | Verification supervision reduces unsupported completion claims | false-success rate | unnecessary test calls |
| H4 | Repair supervision improves recovery after the first failed attempt | recovery success | repeated-action loops |
| H5 | Permission supervision reduces unsafe actions | unsafe action rate | over-refusal / unnecessary approval |
An effect should not be attributed to this dataset unless it survives an appropriate control and ablation analysis.
Quality assurance
The packaged validator checks:
- exact total and split counts;
- exact behavior, record-type, and language quotas;
- required top-level fields;
- unique IDs;
- unique task instructions;
- unique SHA-256 record fingerprints;
- fingerprint integrity;
- split field/file agreement;
- repository-family isolation across splits;
- explicit non-verified provenance;
- expected quality tier.
Run:
python scripts/validate_dataset.py
Regenerate the deterministic release:
python scripts/generate_dataset.py
python scripts/validate_dataset.py --report validation_report.json
python scripts/make_checksums.py
Generation uses the fixed seed 6172026. Identical code and environment should reproduce identical JSONL records and fingerprints.
Data generation methodology
Version 1 is generated from controlled combinations of:
- nine behavioral targets;
- three supervision formats;
- eight language ecosystems;
- seven software-engineering task types;
- three difficulty labels;
- twenty-five synthetic service domains;
- twenty-five component families;
- twenty defect patterns;
- bounded constraint sets;
- behavior-specific chosen and rejected action patterns;
- failure signals and repair sequences.
The generator produces 500 synthetic repository families with 20 tasks per family. Families are assigned to splits before records are written. Each record receives a content-derived SHA-256 fingerprint.
The methodology prioritizes controlled coverage and explicit labels over naturalistic repository realism. This makes the dataset useful for policy experiments and pipeline engineering, while limiting claims about real-world patch correctness.
Biases and limitations
Synthetic regularity
The data is compositionally generated. Models may learn repeated phrasing, common action order, or template artifacts instead of general agent behavior. Mixing with diverse real trajectories and running a template-detection audit are strongly recommended.
No executable repository state
Paths, symbols, commits, tool results, patches, and test commands describe synthetic environments. They are not backed by repository archives in v1. A model trained only on this release may learn the shape of verification without learning reliable execution.
No human preference study
Chosen/rejected pairs encode the dataset designersβ behavioral specification. They do not represent a broad developer preference survey or inter-annotator agreement study.
Tool vocabulary bias
The abstract tool interface favors search/read/patch/test loops. Agents using AST tools, language servers, database consoles, browser tools, distributed execution, or proprietary IDE actions need explicit interface adaptation.
English-only supervision
Task text and annotations are English. The dataset does not establish multilingual agent performance.
Unequal language coverage
Python and TypeScript dominate the mixture. Rust, C#, and C++ are intentionally smaller and should not be used to claim balanced language performance.
Safety coverage is narrow
The safety subset focuses on permission and scope boundaries in coding workflows. It is not a comprehensive cybersecurity, secure-code-generation, or alignment dataset.
Packaged test split is not an independent benchmark
Although repository families are split-disjoint, all records share one generation system. Use an independently sourced executable benchmark for reported model results.
Misuse and out-of-scope use
Do not use this dataset to:
- claim that generated patches pass tests;
- claim human review or production validation;
- benchmark models on the same data used for training;
- train an agent to execute destructive or external actions without runtime authorization controls;
- replace secure sandboxing, access control, audit logging, or human review;
- infer that a structurally valid trajectory is a correct solution to a real repository task.
Any deployed coding agent should enforce permissions and trust boundaries in the runtime. Dataset supervision is not a security boundary.
Roadmap to a Gold release
A future gold_executed tier should require:
- a licensed, materialized repository snapshot;
- an immutable before-state identifier;
- a real task or controlled bug injection;
- recorded tool calls and tool outputs;
- an applicable patch;
- executed targeted tests;
- executed regression tests where feasible;
- static-analysis results;
- a diff-scope audit;
- human review or adjudication;
- contamination and near-duplicate checks;
- an immutable evidence manifest.
Gold should be an evidence tier, not a marketing label.
Project files
JL-AgentBehavior-10K-v1/
βββ README.md
βββ LICENSE.md
βββ CITATION.cff
βββ VERSION
βββ schema.json
βββ stats.json
βββ validation_report.json
βββ checksums.sha256
βββ data/
β βββ train.jsonl
β βββ validation.jsonl
β βββ test.jsonl
βββ docs/
β βββ BEHAVIOR_SPEC.md
β βββ DATA_STATEMENT.md
β βββ EVALUATION_PROTOCOL.md
β βββ SCHEMA.md
β βββ TRAINING_GUIDE.md
βββ scripts/
βββ format_sft.py
βββ generate_dataset.py
βββ make_checksums.py
βββ validate_dataset.py
Versioning
This release follows semantic versioning for the dataset package:
- Patch: documentation, metadata, or validation fixes that do not intentionally change record semantics.
- Minor: additive fields, new compatible supervision views, or additional records.
- Major: schema-breaking changes, changed behavioral taxonomy, or materially different generation/evidence policy.
Current version: 1.0.0.
License
The packaged dataset and original documentation are released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. See LICENSE.md.
Attribution should identify JumpLander and the dataset name/version. Users remain responsible for reviewing the license compatibility of any external data they combine with this release.
Citation
@dataset{jumplander_agentbehavior_10k_2026,
author = {JumpLander},
title = {JL-AgentBehavior-10K: Structured Behavioral Supervision for Coding Agents},
year = {2026},
version = {1.0.0},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/jumplander/JL-AgentBehavior-10K}
}
References
The release is informed by work on interleaved reasoning/action policies, repository-level software-engineering evaluation, and agent-computer interfaces. These references do not imply endorsement of or involvement in this dataset.
- Yao et al. ReAct: Synergizing Reasoning and Acting in Language Models, 2022/2023.
- Jimenez et al. SWE-bench: Can Language Models Resolve Real-World GitHub Issues?, 2023/2024.
- Yang et al. SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering, 2024.
- Hugging Face. Dataset Cards documentation.
About JumpLander
JumpLander is an engineering project focused on AI-assisted software development, coding-agent loops, repository intelligence, specialized training and evaluation datasets, developer tooling, and secure execution infrastructure.
- Website: jumplander.org
- Hugging Face: huggingface.co/jumplander
- GitHub: github.com/jumplander
This dataset is a research artifact under active development. Technical findings, limitations, and reproducible evaluation results are preferred over unsupported capability claims.
- Downloads last month
- 45