# ThinkNEO Control Plane

An open MCP server with 72 tools for AI governance: the ThinkShield defense layer, spend tracking, smart routing and compliance checks.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 3
- Author: thinkneo-ai
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/thinkneo-ai/mcp-server
- Updated: 2026-07-06
- HTML: https://affpapa.org/en/nevibecode/mcp/thinkneo-control-plane

## What it does

ThinkNEO Control Plane plugs into Claude, Codex, Cursor, GitHub Copilot, Windsurf, Cline, Zed and Gemini CLI over HTTP. It ships 72 tools covering guardrails, FinOps, model routing, observability, outcome validation and compliance. ThinkShield sits alongside them as the defense layer: a detection engine plus five rule packs for injection, auth, abuse, recon and header anomalies.

You can run the repo yourself. Clone it, install `requirements.txt` with pip, then start it through `python -m uvicorn src.server:app --host 0.0.0.0 --port 8081`, or bring it up with `docker compose up -d` from the `deploy` folder. The shorter path is the hosted endpoint at `https://mcp.thinkneo.ai/mcp` — the free tier provisions an API key on its own and allows 500 calls a month, with all 72 tools reachable.

What you get here is not the whole ThinkNEO platform. Policy orchestration, the cryptographic audit chain, tenant management and enterprise integrations stay proprietary and run on `thinkneo.ai`, so the open server will not cover a full audit setup. Apache-2.0 means you can fork it and change what you need.

## Requirements
- Remote server: `https://mcp.thinkneo.ai/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http control-plane https://mcp.thinkneo.ai/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add thinkneo-control-plane --url https://mcp.thinkneo.ai/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "thinkneo-control-plane": {
            "url": "https://mcp.thinkneo.ai/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "thinkneo-control-plane": {
            "type": "http",
            "url": "https://mcp.thinkneo.ai/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http thinkneo-control-plane https://mcp.thinkneo.ai/mcp
```

## When to use it
- wire one MCP server into several AI clients at once
- keep an eye on model call spending
- block SQL injection and path traversal attempts at the edge
- validate an agent's output before it reaches the user
- export an audit trail of agent actions

## Example prompts
- Connect the ThinkNEO Control Plane MCP server and list its tools
- Run this request through ThinkShield and tell me if it contains an injection
- Show how many model calls were made today

## Similar
- [Control Plane](https://affpapa.org/en/nevibecode/mcp/control-plane.md) — MCP server · ★ 11
- [Codex Control Plane MCP](https://affpapa.org/en/nevibecode/mcp/codex-control-plane-mcp.md) — MCP server · ★ 124
- [dd-audit-ai-activity](https://affpapa.org/en/nevibecode/skills/dd-audit-ai-activity.md) — skill · 433 installs
- [dd-audit-cost-spike-investigation](https://affpapa.org/en/nevibecode/skills/dd-audit-cost-spike-investigation.md) — skill · 473 installs
- [a2areplayreceipt-mcp](https://affpapa.org/en/nevibecode/mcp/a2areplayreceipt-mcp.md) — MCP server
- [planetscale-traffic-control-recommendations](https://affpapa.org/en/nevibecode/skills/planetscale-traffic-control-recommendations.md) — skill · 1700 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md