grill-with-docs
Pass
Audited by Gen Agent Trust Hub on May 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates entirely within the local development environment. It is designed to read codebase files and documentation templates to ensure architectural alignment. No network operations, external downloads, or third-party dependencies are requested or utilized.
- [SAFE]: No hardcoded credentials, sensitive file access (such as .ssh or .aws paths), or persistence mechanisms were detected in the skill's instructions or supporting templates.
- [PROMPT_INJECTION]: Evaluation of the indirect prompt injection surface (Category 8):
- Ingestion points: The skill explores the local filesystem, specifically files in the
src/directory and root documentation files, to reconcile terminology and verify implementation details against the domain model. - Boundary markers: Absent; the instructions in SKILL.md do not include specific delimiters or directives to ignore potentially malicious instructions embedded within the analyzed codebase files.
- Capability inventory: The agent is empowered to read repository files and write to or update 'CONTEXT.md' and ADR files in the
docs/adr/directory. - Sanitization: Absent; content retrieved from the codebase is used directly to update project documentation. This is an inherent property of its role as a documentation assistant and does not indicate malicious intent.
Audit Metadata