web-oauth

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists solely of markdown documentation outlining methodologies for security researchers to test OAuth implementations. No executable content or automated scripts are provided.\n- [PROMPT_INJECTION]: The content is descriptive and does not include instructions intended to subvert the AI agent's safety guidelines or override its behavioral constraints.\n- [REMOTE_CODE_EXECUTION]: There are no patterns involving package installations, remote script fetching, or dynamic code execution.\n- [DATA_EXFILTRATION]: No sensitive file access, hardcoded credentials, or network transmission logic were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 11:02 PM
Security Audit — agent-trust-hub — web-oauth