doctor

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs structural and semantic validation of local plugin files (manifests, skills, agents) using standard Bash utilities and Python for JSON parsing.
  • [SAFE]: It checks for the presence of required environment variables and credentials (e.g., TAVILY_API_KEY) for MCP servers but does not attempt to read or exfiltrate their values.
  • [SAFE]: External tool recommendations, such as agent-browser from Vercel Labs or the portless utility, target well-known and trusted organizations or services.
  • [SAFE]: The skill uses a platform-native hook mechanism to capture environment snapshots for diagnostics, which is consistent with its stated purpose.
  • [SAFE]: Diagnostic procedures include verifying file integrity and schema compliance for local memory and configuration files, with no evidence of malicious data processing or exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 12:51 AM
Security Audit — agent-trust-hub — doctor