hermespet-macos-ai-companion
Fail
Audited by Snyk on Jun 14, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E005: Suspicious download URL detected in skill instructions.
- Suspicious download URL detected (high risk: 0.70). The set includes an unverified GitHub repo (basionwang-bot/HermesPet) which could distribute unsigned macOS builds if you clone/build or download releases, a generic/placeholder-looking API endpoint (your-gateway.com/v1) that could be attacker-controlled and used to exfiltrate keys or data, and an external site (ara.so) that should be validated — so these sources are potentially risky unless you verify the author, review the code/builds and confirm the gateway domain's trustworthiness.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill's CustomAIService makes runtime HTTP requests to an environment-configured gateway (e.g., HERMES_GATEWAY_BASE_URL="https://your-gateway.com/v1") and POSTs to the /v1/chat/completions endpoint to retrieve model completions which directly control the agent's responses.
Issues (2)
E005
CRITICALSuspicious download URL detected in skill instructions.
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata