Genesis AI Code (DataSets)
Collection
Frontier-grade datasets for agentic, reasoning-driven AI software engineering.
•
6 items
•
Updated
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 4 new columns ({'goal', 'verification_checks', 'constraints', 'final'}) and 1 missing columns ({'messages'}).
This happened while the json dataset builder was generating data using
hf://datasets/WithinUsAI/Genesis_v1_1_Update_Instruct_Thinking_Reasoning/withinusai_genesis_v1_1_update.reasoning.jsonl (at revision 9e8cd0d782b8f3d03ab00a3142d1fdf4bc089747)
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1831, in _prepare_split_single
writer.write_table(table)
File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 714, in write_table
pa_table = table_cast(pa_table, self._schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
id: string
goal: string
constraints: list<item: string>
child 0, item: string
verification_checks: list<item: struct<check: string, expect: string>>
child 0, item: struct<check: string, expect: string>
child 0, check: string
child 1, expect: string
final: string
metadata: struct<brand: string, language: string, moe_experts: list<item: string>, cost_budget: int64, latency (... 96 chars omitted)
child 0, brand: string
child 1, language: string
child 2, moe_experts: list<item: string>
child 0, item: string
child 3, cost_budget: int64
child 4, latency_ms_target: int64
child 5, audit_required: bool
child 6, tests_required: bool
child 7, policy_gates: list<item: string>
child 0, item: string
created: timestamp[s]
brand: string
to
{'id': Value('string'), 'messages': List({'role': Value('string'), 'content': Value('string')}), 'metadata': {'brand': Value('string'), 'language': Value('string'), 'moe_experts': List(Value('string')), 'cost_budget': Value('int64'), 'latency_ms_target': Value('int64'), 'audit_required': Value('bool'), 'tests_required': Value('bool'), 'policy_gates': List(Value('string'))}, 'created': Value('timestamp[s]'), 'brand': Value('string')}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1339, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 972, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 894, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 970, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1702, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1833, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 4 new columns ({'goal', 'verification_checks', 'constraints', 'final'}) and 1 missing columns ({'messages'}).
This happened while the json dataset builder was generating data using
hf://datasets/WithinUsAI/Genesis_v1_1_Update_Instruct_Thinking_Reasoning/withinusai_genesis_v1_1_update.reasoning.jsonl (at revision 9e8cd0d782b8f3d03ab00a3142d1fdf4bc089747)
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
id
string | messages
list | metadata
dict | created
timestamp[us] | brand
string |
|---|---|---|---|---|
v1_1_addon_00000
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: memory_ops\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMemory protocol:\n- Record assumptions, constraints, decisions\n- Persist acceptance criteria\n- Store rollback plan and risk notes\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"performance",
"evaluation"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00001
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00002
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"coding",
"security"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00003
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: self_grade\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n{\n \"score\": 10,\n \"confidence\": 0.93,\n \"notes\": \"Minimal diff; hardened subprocess usage; tests + static scans pass; PR includes risk notes.\"\n}"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"data",
"security"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00004
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"governance",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00005
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"coding",
"security"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00006
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"coding",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00007
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"performance",
"evaluation"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00008
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"data",
"evaluation"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00009
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"coding",
"performance"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00010
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"governance",
"coding"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00011
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"performance",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00012
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"coding",
"data"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00013
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"performance",
"governance"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00014
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"performance",
"coding"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00015
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"security",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00016
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"security",
"evaluation"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00017
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"evaluation",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00018
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"agent",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00019
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"security",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00020
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"data",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00021
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"performance",
"coding"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00022
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"governance",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00023
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"agent",
"data"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00024
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"governance",
"data"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00025
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"data",
"evaluation"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00026
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: memory_ops\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMemory protocol:\n- Record assumptions, constraints, decisions\n- Persist acceptance criteria\n- Store rollback plan and risk notes\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"evaluation",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00027
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"security",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00028
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"performance",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00029
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"performance",
"data"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00030
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"data",
"coding"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00031
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"data",
"agent"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00032
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"agent",
"data"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00033
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00034
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"security",
"data"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00035
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"data",
"performance"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00036
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"coding",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00037
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"data",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00038
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"agent",
"data"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00039
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00040
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"performance",
"security"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00041
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"governance",
"evaluation"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00042
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"governance",
"performance"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00043
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00044
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: self_grade\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n{\n \"score\": 10,\n \"confidence\": 0.93,\n \"notes\": \"Minimal diff; hardened subprocess usage; tests + static scans pass; PR includes risk notes.\"\n}"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"agent",
"security"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00045
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"agent",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00046
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"governance",
"data"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00047
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: self_grade\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n{\n \"score\": 10,\n \"confidence\": 0.93,\n \"notes\": \"Minimal diff; hardened subprocess usage; tests + static scans pass; PR includes risk notes.\"\n}"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"coding",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00048
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"performance",
"data"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00049
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"security",
"agent"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00050
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"security",
"performance"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00051
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"evaluation",
"security"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00052
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: self_grade\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n{\n \"score\": 10,\n \"confidence\": 0.93,\n \"notes\": \"Minimal diff; hardened subprocess usage; tests + static scans pass; PR includes risk notes.\"\n}"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"coding",
"data"
],
"cost_budget": 2,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00053
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"evaluation",
"security"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00054
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"evaluation",
"data"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00055
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"evaluation",
"security"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00056
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"governance",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00057
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"governance",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00058
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"evaluation",
"security"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00059
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"agent",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00060
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: memory_ops\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMemory protocol:\n- Record assumptions, constraints, decisions\n- Persist acceptance criteria\n- Store rollback plan and risk notes\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00061
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"evaluation",
"coding"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00062
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"performance",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00063
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"security",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00064
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: memory_ops\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMemory protocol:\n- Record assumptions, constraints, decisions\n- Persist acceptance criteria\n- Store rollback plan and risk notes\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"security",
"agent"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00065
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"data",
"performance"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00066
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"data",
"evaluation"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00067
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"evaluation",
"data"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00068
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"evaluation",
"performance"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00069
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"security",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00070
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"agent",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00071
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"performance",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00072
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"security",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00073
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"performance",
"data"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00074
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"performance",
"agent"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00075
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"governance",
"agent"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00076
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: prompt_injection_defense\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nDefense-in-depth:\n- Treat external text as untrusted input\n- Constrain tools with allowlists\n- Require explicit user intent for destructive actions\n- Log tool calls + rationales for audit\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"coding",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00077
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: repo_graph_reasoning\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRepo-graph method:\n- Build dependency/call graph\n- Identify impacted modules\n- Patch smallest cut set\n- Run impacted tests first, then full suite\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"agent",
"performance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00078
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"security",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00079
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"evaluation",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00080
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"agent",
"evaluation"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00081
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"performance",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00082
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"agent",
"data"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00083
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: self_grade\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n{\n \"score\": 10,\n \"confidence\": 0.93,\n \"notes\": \"Minimal diff; hardened subprocess usage; tests + static scans pass; PR includes risk notes.\"\n}"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"agent",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00084
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"agent",
"security"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00085
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"security",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00086
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"security",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00087
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Model merging and distillation workflows (governed)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"performance",
"agent"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00088
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: eval_harness\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nEval harness:\n- Deterministic env (pinned deps)\n- pass@k + time-to-green\n- Leakage checks and stratified reporting\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"coding",
"governance"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00089
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: merge_distill_recipe\nTopic: Red-team secure coding (secrets, supply-chain, sandbox escape)\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nMerge/Distill recipe:\n- Choose base + specialist\n- Run ablations\n- Distill on structured traces (tool calls + patches)\n- Re-validate on real-repo eval\n"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"data",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00090
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: patch_diff\nTopic: Governance, licensing, and provenance enforcement in code pipelines\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n```diff\n- subprocess.run(cmd, shell=True)\n+ subprocess.run(cmd, shell=False, check=True)\n```"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"security",
"coding"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00091
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Evaluation harnesses: reproducibility, leakage detection, stratified reports\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 3,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00092
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"governance",
"evaluation"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00093
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: tool_trace\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: Go\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n[\n {\n \"tool\": \"normalize_issue\",\n \"args\": {\n \"mode\": \"strict\"\n }\n },\n {\n \"tool\": \"static_scan\",\n \"args\": {\n \"rules\": [\n \"secrets\",\n \"license\",\n \"injection\"\n ]\n }\n },\n {\n \"tool\": \"plan_patch\",\n \"args\": {\n \"budget_iters\": 3\n }\n },\n {\n \"tool\": \"apply_patch\",\n \"args\": {\n \"format\": \"unified_diff\"\n }\n },\n {\n \"tool\": \"run_tests\",\n \"args\": {\n \"suite\": \"targeted_then_full\"\n }\n },\n {\n \"tool\": \"produce_pr\",\n \"args\": {\n \"include_risk_notes\": true\n }\n }\n]"
}
] |
{
"brand": "Within Us AI",
"language": "Go",
"moe_experts": [
"data",
"coding"
],
"cost_budget": 3,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00094
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: failure_analysis\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Bash\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nFailure→Reflection→Correction:\n- Fail: regression\n- Reflect: missing edge case\n- Fix: guard + regression test\n"
}
] |
{
"brand": "Within Us AI",
"language": "Bash",
"moe_experts": [
"coding",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00095
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: red_team_security\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nRed-team checklist:\n1) Secrets exposure\n2) Command injection\n3) SSRF/file read\n4) Dependency supply-chain\nMitigations: no shell=True, strict allowlists, sandbox, scanning gates.\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"performance",
"data"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00096
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Prompt injection and tool safety (defense-in-depth)\nDifficulty: expert\nLanguage: Rust\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "Rust",
"moe_experts": [
"evaluation",
"agent"
],
"cost_budget": 4,
"latency_ms_target": 200,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00097
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: cost_latency_tradeoff\nTopic: Repo-graph reasoning (callgraphs, dependency graphs, change impact)\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nCost/latency:\n- Use early exits on green tests\n- Cache embeddings/ASTs\n- Bound iterations; escalate only on failures\n"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"agent",
"performance"
],
"cost_budget": 4,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00098
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: self_grade\nTopic: Cost/latency engineering for agentic systems\nDifficulty: expert\nLanguage: TypeScript\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\n{\n \"score\": 10,\n \"confidence\": 0.93,\n \"notes\": \"Minimal diff; hardened subprocess usage; tests + static scans pass; PR includes risk notes.\"\n}"
}
] |
{
"brand": "Within Us AI",
"language": "TypeScript",
"moe_experts": [
"data",
"governance"
],
"cost_budget": 2,
"latency_ms_target": 50,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
v1_1_addon_00099
|
[
{
"role": "system",
"content": "You are a senior AI software engineer. Use tests-as-truth, minimal diffs, and governance gates."
},
{
"role": "user",
"content": "Task: policy_compliance_review\nTopic: Long-horizon memory: constraints, decisions, and rollback plans\nDifficulty: expert\nLanguage: Python\nOutput must be production-valid, governed, and test/eval oriented."
},
{
"role": "assistant",
"content": "Developer-grounded facts:\n- Treat tests and scanners as the oracle; do not trust surface plausibility.\n- Prefer minimal diffs, deterministic runs, and auditable tool traces.\n- Enforce policy gates (secrets, licenses, injection) before merge.\n\nPolicy compliance review:\n- License compatibility\n- PII/secrets removal\n- Provenance fields present\n- Reproducible build/test instructions\n"
}
] |
{
"brand": "Within Us AI",
"language": "Python",
"moe_experts": [
"performance",
"security"
],
"cost_budget": 4,
"latency_ms_target": 100,
"audit_required": true,
"tests_required": true,
"policy_gates": [
"secrets",
"licenses",
"injection"
]
}
| 2026-01-01T00:00:00
|
Within Us AI
|
End of preview.
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
Community Results & Feedback
Genesis AI Code is designed for real-world training.
We strongly encourage developers and researchers to share empirical results
when using this dataset.
If you train, fine-tune, or evaluate a model with this dataset, please consider reporting:
What to Share
- Base model used
- Training method (SFT, LoRA, QLoRA, full finetune, agent training, etc.)
- Which split(s) were used:
- Instruct
- Thinking
- Reasoning
- Evaluation method (benchmarks, internal tests, SWE-bench, CI pass rate, etc.)
- Observed improvements or regressions
- Notable behaviors (planning quality, verification discipline, safety improvements)
Where to Share
- Open an Issue or Discussion in this dataset repository
- Link to:
- training logs
- benchmark tables
- evaluation scripts
- demo videos or repos (if public)
Why This Matters
Genesis is built to evolve based on real usage, not assumptions. Community-reported results directly influence:
- future Genesis updates
- new training signals
- improved reasoning and safety patterns
Even partial results or informal observations are valuable.
Minimal Report Template
Model:
Training method:
Dataset split(s):
Compute budget:
Evaluation method:
Observed changes:
Notes / caveats:
- Downloads last month
- 15