Cordon is an MCP proxy gateway that sits between an AI agent and one or more MCP servers to enforce per-tool policies, approval workflows, and audit logging. It intercepts tool calls from clients like Claude Desktop and routes them through policy checks before they reach the target MCP servers.
From the registry: Cordon for MCP. Security gateway with policy enforcement, audit log, and HITL approvals.
$ npx cordon-cli initPlease install the `cordon` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Node.js and npm** — Node.js with npm is required to run the CLI via npx or install it globally.
- **Initialize Cordon in your project** — Run the init command in the project where your MCP client config exists. This generates cordon.config.ts and patches the client config to route tool calls through Cordon. Run: `npx cordon-cli init`
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["cordon-cli","start"]
Note: Cordon is an aggregating MCP proxy that sits between the MCP client and your underlying MCP servers. The README indicates current support is stdio-based and explicitly notes HTTP/SSE transport support is planned, not yet available. Configuration of upstream servers and policies is done in cordon.config.ts rather than through MCP env vars shown in the README.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Dead code, security, secrets detection and code quality for Python, TypeScript, Go.