interlock-mcp

Zigerus/interlock-mcp

★ 0 stars Python AI/LLM Updated 9d ago

A human-in-the-loop governance interlock for AI agents (MCP server): agents propose changes, a human countersigns the exact plan, then it executes stage-by-stage — precondition-checked, verified, and audited. Bring your own executors.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "interlock-mcp": {
      "command": "uvx",
      "args": [
        "interlock-mcp"
      ]
    }
  }
}

Or install with pip: pip install interlock-mcp

README Excerpt

**A human-in-the-loop governance interlock for AI agents (MCP server).** Agents *propose* changes; a human *countersigns the exact plan*; only then does it *execute* — stage by stage, precondition-checked against live state, verified, and audited. Bring your own executors. > An *interlock* is a safety mechanism that prevents an action until every required

interlock-mcp

Quick Install

README Excerpt

Related AI/LLM Servers