grill-with-docs

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified. The skill is designed to operate within the local project context to facilitate design discussions and documentation updates.
  • [NO_CODE]: The skill is composed entirely of natural language instructions and documentation templates. It contains no executable scripts, binaries, or configuration files that trigger code execution.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads untrusted content from the user's codebase and existing documentation files. However, this is assessed as safe/low risk because the skill's capabilities are limited to conversation and generating markdown documentation.
  • Ingestion points: Codebase exploration and reading local documentation files (CONTEXT.md, ADRs).
  • Boundary markers: None specified for ingested content.
  • Capability inventory: File reading, file writing (markdown documentation), and codebase searching.
  • Sanitization: No explicit sanitization or escaping of ingested data before interpolation into the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 07:05 AM
Security Audit — agent-trust-hub — grill-with-docs