resolving-merge-conflicts
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest untrusted data from git commit messages, PR descriptions, and issue trackers to understand change intent. This creates a surface where malicious instructions could be embedded in the project metadata.\n
- Ingestion points: External data sources including git history, commit messages, PRs, and issues/tickets mentioned in SKILL.md.\n
- Boundary markers: The instructions do not specify any delimiters or safety warnings to help the agent distinguish between data and potentially malicious instructions within these sources.\n
- Capability inventory: The skill involves executing git commands and running local automated checks like tests and typecheckers.\n
- Sanitization: There are no instructions to sanitize or validate the content retrieved from external sources.
Audit Metadata