grill-with-docs

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions focus on interviewing the user to refine project terminology and documentation. It follows standard software engineering practices for Decision Records (ADR) and Domain-Driven Design (DDD) glossaries.
  • [COMMAND_EXECUTION]: The skill uses natural language instructions to interact with the codebase (e.g., 'explore the codebase', 'update linear'). These are standard agentic behaviors for development tools and do not involve arbitrary shell injection or unauthorized command execution.
  • [DATA_EXPOSURE]: While the skill reads sensitive-sounding filenames like CONTEXT.md and docs/adr/, these are standard project documentation files intended for the agent's context to maintain consistency. No credentials or private keys are targeted.
  • [PROMPT_INJECTION]: The skill includes instructions to 'Interview me relentlessly', which is a stylistic persona instruction for the agent to ensure thoroughness, not a safety bypass or an attempt to override system constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 02:27 PM
Security Audit — agent-trust-hub — grill-with-docs