grilling

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs standard file system operations including creating, writing, and deleting temporary files (.grill-tree.md) and UI mockups. It also instructs the agent to modify the .gitignore file to ensure scratch state is not committed, which is a standard development practice.
  • [DATA_EXPOSURE]: The skill involves exploring the local codebase to answer design questions. This access is limited to the local environment and is necessary for the skill's stated purpose of stress-testing plans and designs.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided plans and design documents and persists decisions to project files (DESIGN.md, CONTEXT.md, ADRs). While this is an ingestion surface for untrusted data, the risk is mitigated because the content is treated as documentation rather than executable code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 06:00 AM
Security Audit — agent-trust-hub — grilling