grilling
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external inputs to generate interrogation questions and project documentation.
- Ingestion points: The skill reads from
userPrompt,wayfinderBrief, and existing root documents (AGENTS.md,CLAUDE.md) to extract requirements and identify risk areas (SKILL.md). - Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between user input and system instructions during the analysis phase.
- Capability inventory: The skill has the capability to write and append data to local project files, specifically
walkthrough.md(SKILL.md). - Sanitization: No evidence of input validation, filtering, or sanitization is present before external content is incorporated into the conversation context or persistent files.
Audit Metadata