mem-automation
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill instructs the agent to call RUBE_SEARCH_TOOLS and RUBE_GET_TOOL_SCHEMAS against the external Rube MCP/Composio endpoint (https://rube.app/mcp / composio.dev) and to read the returned tool schemas and recommended execution plans, which are third-party, untrusted content the agent must interpret at runtime.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill requires contacting the external MCP endpoint https://rube.app/mcp at runtime to discover and invoke RUBE_* tools (RUBE_SEARCH_TOOLS, RUBE_MULTI_EXECUTE_TOOL), which can deliver tool schemas and execute remote actions that directly control agent behavior.
Audit Metadata