discover-proxies

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No malicious instructions or attempts to bypass agent safety filters were detected.
  • [Data Exposure & Exfiltration] (SAFE): The skill does not access sensitive files, hardcode credentials, or initiate network connections.
  • [Indirect Prompt Injection] (LOW): This skill directs the agent to ingest additional data from the local file system (<cc-polymath-root>/skills/proxies/). This increases the attack surface if those files contain malicious instructions, but is the standard behavior for a gateway skill.
  • Ingestion points: SKILL.md (references local markdown files)
  • Boundary markers: None provided in the gateway file
  • Capability inventory: File-reading via the Read command
  • Sanitization: Not applicable (standard documentation loading)
  • [Unverifiable Dependencies] (SAFE): No external packages (npm/pip) or remote scripts are downloaded or executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:51 PM