auto-grill
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data, such as
PLAN.md, issues, or PRDs, which could contain malicious instructions. - Ingestion points: The agent is instructed to find and 'grill' plans that already exist in the conversation context or codebase (e.g.,
PLAN.md, issues, PRD). - Boundary markers: The skill lacks explicit instructions to treat the ingested plan content as data only, or to ignore any embedded directives that might attempt to hijack the agent's logic during the decision-making loop.
- Capability inventory: The agent possesses the capability to explore the codebase and rewrite files (the 'Resolved plan') based on decisions made during the loop.
- Sanitization: No sanitization or validation logic is specified for the external content before it is incorporated into the agent's reasoning process.
Audit Metadata