claude-code-operator
Warn
Audited by Snyk on Mar 3, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill instructs the agent to call and list MCP tools from a third-party MCP server (e.g., https://mcp-on-edge.edgeone.app/mcp-server shown in "Workflow 1" and "Workflow 3"), which the agent must read/interpret to decide and invoke tools — exposing it to untrusted external content that can influence actions.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill issues runtime curl POSTs to https://mcp-on-edge.edgeone.app/mcp-server (JSON-RPC methods like "tools/call" and "tools/list") which invoke MCP tools (e.g., deploy-html), so external server-side code is executed at runtime and relied upon for deployment workflows.
Audit Metadata