systematic-debugging
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from error logs and web content, creating an attack surface for instructions embedded in external data. Ingestion points: Error messages and multi-component logs (SKILL.md). Boundary markers: The skill explicitly requires treating external data as hostile (SKILL.md). Capability inventory: Access to terminal execution via the ExecutionProxy interface and the find-polluter.sh script. Sanitization: Redaction of secrets and PII is mandated before sharing context with subagents.
- [COMMAND_EXECUTION]: The utility script find-polluter.sh executes npm test on files matching a search pattern, representing a functional shell capability used for test bisection.
- [DYNAMIC_EXECUTION]: The debugging workflow requires automated code execution via rtk npm test to verify code integrity at multiple gates (Unit, Contract, E2E) as part of the 'Iron Law' gating ladder.
Audit Metadata