ai-data-readiness-triage
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The
scripts/quick_validate_skill.pyscript usessubprocess.runto execute the project's internal test suite (tests/test_skill_contract.py). This is a local validation tool for developers and does not pose a threat to the execution environment. - [REMOTE_CODE_EXECUTION]: The
compile()function is utilized intests/test_skill_contract.pyas a linting mechanism to verify the syntax of Python files within the skill package. It does not execute untrusted external code. - [DATA_EXFILTRATION]: Although the skill is designed to profile and analyze local data such as CSV files, schemas, and reports, it includes explicit instructions in
SKILL.mdandreferences/provider-interop.mdto prevent the unauthorized upload of private data to external services. - [PROMPT_INJECTION]: The skill provides adversarial test fixtures in
references/acceptance-tests.mdandtests/fixtures/adversarial.md. These are used to benchmark the agent's ability to resist pressure to overclaim readiness and to identify missing information correctly.
Audit Metadata