document

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs project-wide filesystem scans and evaluates git commit history to determine project maturity and documentation coverage.
  • [DATA_EXFILTRATION]: The agent is instructed to read various project configuration files and environment templates. This behavior is localized for auditing purposes and does not involve unauthorized data transmission.
  • [PROMPT_INJECTION]: The skill processes untrusted project data to generate audit reports, representing a surface for indirect prompt injection.
  • Ingestion points: SKILL.md specifies scanning project configuration files, documentation artifacts, and git history.
  • Boundary markers: The instructions do not define specific markers to isolate or ignore instructions potentially embedded within the audited files.
  • Capability inventory: The skill can read project files and append data to a local telemetry log located at ~/.claude/projects/skill-telemetry.md.
  • Sanitization: No mechanisms for sanitizing or escaping the content of scanned files before inclusion in the health report are mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:58 AM
Security Audit — agent-trust-hub — document