{
  "markdown": "# ThinkNEO MCP Server\n\n> Open MCP server with built-in defense layer (ThinkShield).\n> Part of the [ThinkNEO Platform](https://thinkneo.ai) — enterprise AI governance.\n\n[![License: Apache-2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)\n[![MCP Protocol](https://img.shields.io/badge/MCP-2024--11--05-green.svg)](https://modelcontextprotocol.io)\n[![A2A Protocol](https://img.shields.io/badge/A2A-v0.3.0-blue.svg)](https://github.com/google/A2A)\n[![Tools](https://img.shields.io/badge/Tools-72-orange.svg)](#mcp-tools)\n[![A2A Skills](https://img.shields.io/badge/A2A_Skills-24-blueviolet.svg)](#a2a-capabilities)\n[![Glama AAA](https://img.shields.io/badge/Glama-AAA-brightgreen.svg)](https://glama.ai/mcp/servers/@thinkneo-ai/mcp-server)\n[![Tests](https://github.com/thinkneo-ai/mcp-server/actions/workflows/tests.yml/badge.svg)](https://github.com/thinkneo-ai/mcp-server/actions/workflows/tests.yml)\n\n---\n\n## What This Is\n\nAn open-source MCP server providing 72 tools for AI governance, observability, and security:\n\n- **ThinkShield** — Production defense layer: detection engine, 5 rule packs, runtime middleware. 145 tests, p99 < 1ms.\n- **72 MCP Tools** — Governance, guardrails, FinOps, smart routing, observability, compliance, outcome validation, and more.\n- **24 A2A Skills** — Bidirectional MCP-A2A bridge for Google's Agent-to-Agent Protocol (v0.3.0, Linux Foundation).\n- **Apache-2.0** — Use it, fork it, contribute.\n\n## What This Is Not\n\n- **Not the full ThinkNEO Platform.** Governance orchestration, cryptographic audit chain, tenant management, and enterprise integrations are proprietary and run at [thinkneo.ai](https://thinkneo.ai).\n- **Not a standalone security product.** ThinkShield is the defense component of a larger governed platform.\n\n## Why Open\n\nWe open-source our defense layer because real security doesn't depend on hidden rules — it depends on tested, audited, continuously improved detection plus a strong governance moat around it.\n\nSnort. Suricata. Falco. OWASP CRS. The security industry runs on open detection. We follow that tradition.\n\nThe detection is open. The governance is proprietary. That's where the moat is.\n\n## Architecture\n\n```\n                        Open Source (this repo)                    Proprietary (thinkneo.ai)\n                   ┌─────────────────────────────────┐     ┌──────────────────────────────────┐\n                   │                                 │     │                                  │\n  MCP Clients ────>│  72 MCP Tools                   │     │  Governance Orchestration         │\n  (Claude, Cursor, │  ├── Guardrails & Safety        │────>│  ├── Policy Engine (AIRGP)        │\n   ChatGPT, etc.)  │  ├── FinOps & Smart Routing     │     │  ├── Cryptographic Audit Chain    │\n                   │  ├── Observability              │     │  ├── Tenant Management            │\n  A2A Agents ─────>│  ├── Compliance & Validation    │     │  ├── Enterprise Integrations      │\n  (Google A2A)     │  └── MCP-A2A Bridge (24 skills) │     │  └── SLA & Support                │\n                   │                                 │     │                                  │\n                   │  ThinkShield Defense Layer       │     │  SHA-256 Hash Chain (949K+ rows)  │\n                   │  ├── Detection Engine            │     │  Stripe Billing                   │\n                   │  ├── 5 Rule Packs               │     │  Resend Email                     │\n                   │  └── ASGI Middleware             │     │  Multi-tenant Auth                │\n                   │                                 │     │                                  │\n                   └─────────────────────────────────┘     └──────────────────────────────────┘\n                          Apache-2.0 License                      Commercial License\n```\n\n## Quickstart\n\n```bash\n# Clone\ngit clone https://github.com/thinkneo-ai/mcp-server.git\ncd mcp-server\n\n# Install\npip install -r requirements.txt\n\n# Run\npython -m uvicorn src.server:app --host 0.0.0.0 --port 8081\n\n# Test\npython -m pytest tests/ -q\n```\n\nOr with Docker:\n\n```bash\ncd deploy\ndocker compose up -d\n```\n\nConnect from Claude Desktop, Cursor, or any MCP client:\n```\nhttps://mcp.thinkneo.ai/mcp\n```\n\nFree tier: 500 calls/month, auto-provisioned API key. All 72 tools available.\n\n## Components\n\n| Directory | Description | License |\n|-----------|-------------|---------|\n| `src/tools/` | 72 MCP tools — governance, security, FinOps, observability | Apache-2.0 |\n| `src/thinkshield/` | Defense layer — detection engine, 5 rule packs | Apache-2.0 |\n| `tests/thinkshield/` | ThinkShield test suite — 145 tests + attack/benign fixtures | Apache-2.0 |\n| `agent.json` | A2A Agent Card — 24 skills bridged from MCP | Apache-2.0 |\n\n### ThinkShield Rule Packs\n\n| Pack | Detects |\n|------|---------|\n| `injection` | SQL injection, XSS, command injection, path traversal |\n| `auth` | Credential stuffing, brute force, token replay, privilege escalation |\n| `abuse` | Rate abuse, resource exhaustion, API scraping |\n| `recon` | Path probing, tool enumeration, method probing, fingerprinting |\n| `headers` | Header anomalies, spoofing, missing security headers |\n\n### MCP Tools (72)\n\nGovernance (6) | Guardrails (3) | FinOps (4) | Smart Router (4) | Trust Score (2) | Registry (5) | Bridge (4) | Observability (5) | Business Value (6) | A2A Control (4) | Optimization (1) | Outcome Validation (4) | Policy Engine (4) | Benchmarking (3) | Compliance (2) | Agent SLA (4) | Audit Export (3) | Cache (3) | Security (5) | Tokens (1) | Memory (2) | Scheduling (1) | Alerts (1)\n\nFull tool reference: [docs/quickstart.md](docs/quickstart.md)\n\n## MCP Spec Compliance\n\nComplete Model Context Protocol 2024-11-05 implementation. Forward-compatible with MCP 2025-03-26.\n\n| Capability | Status | Details |\n|------------|--------|---------|\n| tools | 72 tools, full annotations | destructiveHint, readOnlyHint, idempotentHint, openWorldHint |\n| resources | 2 resources | Getting Started guide, Supported Providers |\n| prompts | 2 prompts with completions | governance_audit, policy_preflight |\n| logging | logging/setLevel | 8 levels, per-session, audit trail |\n| completions | completion/complete | workspace (auth-scoped), provider, model (provider-aware) |\n\n## Ecosystem\n\nThis repo is part of the ThinkNEO ecosystem:\n\n| Project | Description |\n|---------|-------------|\n| [ThinkNEO Platform](https://thinkneo.ai) | Enterprise AI governance platform |\n| [AIRGP](https://airgp.space) | AI Runtime Governance Protocol — open standard |\n| [A2ASTC](https://a2astc.space) | A2A Security & Trust Conformance |\n| [ThinkNEO SMB Hub](https://thinkneo.app) | Business applications for SMBs |\n| [Robotics Governance](https://robots.thinkneo.pro) | Robot fleet governance dashboard |\n\n## Security\n\nSee [SECURITY.md](SECURITY.md) for vulnerability reporting.\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md).\n\n## Related\n\n| Server | Description | Tools |\n|--------|-------------|-------|\n| [thinkneo-control-plane](https://glama.ai/mcp/servers/thinkneo-ai/mcp-server) | Enterprise AI Control Plane (this repo) | 72 tools |\n| [thinkneo-mcp-smb-products](https://glama.ai/mcp/servers/thinkneo-ai/mcp-smb-products) | SMB standalone products — self-serve via TNC credits | 37 tools |\n\n## License\n\nApache-2.0 — see [LICENSE](LICENSE).\n\n## About\n\n**ThinkNEO AI Technology Co., Ltd.** — Hong Kong CR No. 2296774.\n\nBuilt by the team behind the ThinkNEO Enterprise AI Control Plane, AIRGP protocol, and A2ASTC conformance suite.\n\n[NVIDIA Inception](https://nvidia.com/inception) | [Anthropic Partner](https://anthropic.com) | [Linux Foundation A2A](https://github.com/google/A2A)\n",
  "bytes": 7678,
  "sha": "1b78dc2cf4262fd55cddd42db56afaf7ddb1b75e103f732f7bb45802f4f587bf",
  "repo_slug": "thinkneo-ai/mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_ai_thinkneo_control_plane_692b1ca2/readme"
}