grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository codebase and existing documentation files during its exploration phase.
- Ingestion points: Instructions in SKILL.md direct the agent to "explore the codebase" and read files like CONTEXT.md, CONTEXT-MAP.md, and ADRs to resolve dependencies and terminology.
- Boundary markers: The instructions lack explicit delimiters or safety instructions to prevent the agent from obeying instructions that might be embedded within the files being analyzed.
- Capability inventory: The skill utilizes codebase exploration (read) and documentation updates (write) to CONTEXT.md and sequential ADR files. It does not perform network operations or arbitrary shell command execution.
- Sanitization: There is no evidence of sanitization, filtering, or validation of the content read from the codebase before it is used to influence the agent's logic or documentation writes.
Audit Metadata