codebase-cleanup-refactor-clean
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill involves the analysis and modification of user-provided codebases, creating a surface where malicious instructions embedded in the source code could influence the agent's behavior.\n
- Ingestion points: The agent processes user code through the $ARGUMENTS parameter defined in SKILL.md.\n
- Boundary markers: There are no explicit instructions or delimiters provided to prevent the agent from following natural language instructions found within comments or strings in the processed code.\n
- Capability inventory: The skill allows for identifying refactor candidates and applying changes, which involves file-write operations in an integrated environment.\n
- Sanitization: The skill lacks explicit sanitization or filtering steps for the ingested code to mitigate injection risks.\n- [EXTERNAL_DOWNLOADS]: The playbook contains configuration templates that reference established GitHub Actions and third-party code review services.\n
- Evidence: References to official actions such as actions/checkout@v4, github/copilot-autofix@v1, and github/codeql-action/analyze@v3 are found in the CI/CD workflow examples within resources/implementation-playbook.md. It also references well-known services like coderabbitai and codiumai for automated code reviews.
Audit Metadata