openclaw-integration

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • SAFE (SAFE): No malicious behaviors or patterns detected. The skill provides configuration instructions for a legitimate service.
  • EXTERNAL_DOWNLOADS (LOW): The skill configures a connection to the official Genfeed.ai MCP server (https://mcp.genfeed.ai/mcp) and references its CLI tool. These external references are standard for this type of integration and align with the skill's primary purpose.
  • CREDENTIALS_UNSAFE (SAFE): Authentication is handled via API keys, but the skill correctly uses placeholders like 'YOUR_API_KEY' and 'gf_live_xxx' in its examples, directing users to the official platform for key generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:48 PM