agenty-automation
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires connecting to the Rube MCP endpoint (https://rube.app/mcp) at runtime—calls like RUBE_SEARCH_TOOLS and RUBE_MULTI_EXECUTE_TOOL fetch tool schemas and execution plans that directly determine the agent's instructions/actions, making this external content a required runtime dependency that can control prompts/execute code.
Audit Metadata