grill-with-docs

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were identified in the skill instructions or associated scripts.
  • [COMMAND_EXECUTION]: The skill incorporates three Python scripts (context_md_linter.py, adr_scanner.py, and glossary_code_consistency.py) designed to validate documentation and cross-reference terms with the codebase. These scripts are self-contained, utilize only the Python standard library, and do not perform network requests or execute arbitrary commands.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted local data (codebase files and markdown documentation), the analysis is restricted to static regex matching and frequency counting. The risk of the agent obeying instructions embedded in these files is low and inherent to developer-focused tools that read code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 09:17 PM
Security Audit — agent-trust-hub — grill-with-docs