rigor
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user requests and repository content, which creates a vulnerability surface where instructions embedded in external files could influence the agent.
- Ingestion points: As defined in
SKILL.md, the agent resolving a task must ingest the 'request and repository context.' - Boundary markers: The skill emphasizes evidence-based verification but does not specify technical delimiters to separate untrusted repository content from system instructions.
- Capability inventory: The agent has the ability to search the repository, execute shell commands, and modify files as described in the
bug-fix.md,feature.md, andrefactoring.mdplaybooks. - Sanitization: No explicit procedures for sanitizing or escaping untrusted data from the repository are included in the instructions.
Audit Metadata