clean-code
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted source code for review and refactoring, which presents an attack surface for indirect prompt injection. * Ingestion points: Source code files and pull request data analyzed for clean code violations (SKILL.md). * Boundary markers: Structured 'Clean Code Review' report format used to separate agent findings from the analyzed content (SKILL.md). * Capability inventory: The skill instructs the agent to run project test suites ('Run the suite. Green? Refactor.') which involves execution of potentially untrusted project code (SKILL.md). * Sanitization: Absent; the skill relies on the agent's default safety guardrails for handling untrusted input.
- [COMMAND_EXECUTION]: The skill includes an integrity validation script
eval/run_eval.pyintended for self-testing. The script uses the Python standard library to perform local file read and write operations within the skill's own directory and system temporary directories. It does not perform network operations or access sensitive environment variables.
Audit Metadata