An MCP server for pre-flight auditing of payment transactions against 9 security checks, including token integrity, budget limits, allowlist/category restrictions, expiry, rate limits, replay, and revocation. It can also revoke tokens, clear nonces, verify audit hash chains, and simulate common payment attack vectors.
From the registry: Security guardrails for AI agent payments

$ pip install mcpPlease install the `agentpay-sentinel-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Python 3.10+** — Python 3.10 or newer is required to run the MCP server.
- **Install the mcp Python package** — Install the required Python MCP package before running the server. Run: `pip install mcp`
- **Obtain the server source code** — The configuration runs the server via a local Python file path, so you need a local checkout of the repository and must know the absolute path to server.py. (https://github.com/Rumblingb/agentpay-sentinel-mcp)
Canonical MCP server config (stdio transport):
- command: `python`
- args: ["/absolute/path/to/agentpay-sentinel-mcp/server.py"]
Note: The README only documents running the server locally via Python using a path to server.py. No environment variables are described. Server state persists under ~/.sentinel/.
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.