grill-with-docs

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is focused on documentation and domain modeling. It assists in maintaining a project's glossary and decision records by reading existing files and proposing updates.
  • [PROMPT_INJECTION]: No malicious injection patterns or attempts to bypass safety filters were found. The use of instructional language like "sabatine sem dó" (grill me without mercy) is context-appropriate for a stress-testing persona.
  • [DATA_EXFILTRATION]: No network operations, exfiltration patterns, or unauthorized data access attempts were detected. The skill interacts exclusively with project documentation and source code for analysis.
  • [REMOTE_CODE_EXECUTION]: No dynamic code execution, remote script downloads, or subprocess calls were found. The skill's operations are limited to reading and writing markdown-based documentation.
  • [CREDENTIALS_UNSAFE]: No hardcoded secrets or attempts to access sensitive configuration files (e.g., .env, credentials files) were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 05:24 PM
Security Audit — agent-trust-hub — grill-with-docs