refactor
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external code provided by the user (PRs, functions, files), which acts as an untrusted data source. This introduces a surface for indirect prompt injection where instructions embedded in the code being refactored could potentially influence the agent's actions during the refactoring or committing process.
- Ingestion points: User-provided code snippets, files, and pull requests mentioned in the Workflow section.
- Boundary markers: None specified; the skill lacks explicit instructions to treat comments or strings within the code as non-instructional data.
- Capability inventory: The skill expects the agent to perform file modifications and Git operations (separate commits, generating diffs, and linking to commits).
- Sanitization: No sanitization or safety filtering of the input code is defined.
Audit Metadata