grill
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze repository content, including files, documentation, and commit history, to inform its design recommendations. This establishes an indirect prompt injection surface where malicious instructions embedded in a project's source code or documentation could attempt to influence the agent's brainstorming logic or design decisions.
- Ingestion points: The workflow explicitly starts with 'Explore context — files, docs, recent commits relevant to the idea' in
SKILL.md. - Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore potentially malicious embedded prompts within the analyzed repository data.
- Capability inventory: While this skill focuses on logic and planning without direct execution or network capabilities, it serves as a gatekeeper for subsequent implementation skills (
write-spec,write-plan,tdd). - Sanitization: No explicit sanitization or validation of the ingested repository content is mentioned.
Audit Metadata