grill-softly
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external repository data to generate design documents and glossaries, creating a potential surface for indirect prompt injection.
- Ingestion points: Processes repository branches and evidence as described in the skill metadata (SKILL.md).
- Boundary markers: No explicit delimiters or instructions to ignore embedded prompts are present in the provided instructions.
- Capability inventory: The skill body specifies calling two other skill tools: 'torture-gently' and 'domain-modeling' (SKILL.md).
- Sanitization: There is no evidence of sanitization or validation of the ingested repository content before it is processed by the agent.
Audit Metadata