Documentation

Complete guide to integrating Centre AI with your AI assistant.

Quick Start

Get Centre AI running in minutes:

1. Create Account: Sign up at centre-ai.de
2. Generate API Key: Go to Dashboard → API Keys → Create New Key
3. Configure Client: Add the configuration to your MCP client
4. Start Using: Your AI now has persistent memory!

Claude Desktop Setup

The most common way to use Centre AI with Claude Desktop:

Step 1: Create API Key

1. Log in to centre-ai.de
2. Open your Dashboard
3. Click "Manage API Keys"
4. Create a new key with a descriptive name (e.g., "Claude Desktop")
5. Important: Copy the key immediately - it's only shown once!

Step 2: Configure Claude Desktop

1. Open Claude Desktop Settings
2. Navigate to: Settings → Developer → Edit Configuration
3. Open claude_desktop_config.json
4. Add this configuration:

{
  "mcpServers": {
    "centre-ai": {
      "command": "npx",
      "args": [
        "-y",
        "supergateway",
        "--sse",
        "https://centre-ai.de/mcp/sse"
      ],
      "env": {
        "CENTRE_AI_API_KEY": "cai_YOUR_API_KEY_HERE"
      }
    }
  }
}

Step 3: Test Connection

1. Restart Claude Desktop
2. Open a new chat
3. You should see Centre AI tools (e.g., "create_memory", "get_memory")
4. Test with: "Save a memory: I like coffee"

Claude Web (OAuth)

Connect Centre AI to Claude.ai directly via OAuth - no API key needed:

Step 1: Open Claude.ai

1. Go to claude.ai
2. Sign in with your Anthropic account
3. Open Settings (gear icon)

Step 2: Add MCP Server

1. Navigate to "Integrations" or "MCP Servers"
2. Click "Add MCP Server"
3. Enter Server URL: https://centre-ai.de
4. Click "Connect"

Step 3: Authorize

1. You'll be redirected to Centre AI
2. Log in with your Centre AI account (or register)
3. Approve the access request
4. You'll be redirected back to Claude.ai

ChatGPT Integration

Centre AI can also be used with ChatGPT:

Option 1: Custom GPT (Recommended)

1. Create an API Key in your Centre AI Dashboard
2. Go to ChatGPT GPT Editor
3. Create a new Custom GPT
4. Under "Actions", add the Centre AI API:

OpenAPI Schema URL: https://centre-ai.de/api/openapi.json

Authentication:
- Type: API Key
- Header: X-API-Key
- Value: cai_YOUR_API_KEY_HERE

Other MCP Clients

Centre AI works with all MCP-compatible clients:

Universal Configuration

For all MCP clients, use this base configuration:

Server URL: https://centre-ai.de/mcp/sse
Transport: SSE (Server-Sent Events)
Header: X-API-Key: cai_YOUR_API_KEY_HERE

# Or with supergateway (if client doesn't support SSE):
Command: npx -y supergateway --sse https://centre-ai.de/mcp/sse
Env: CENTRE_AI_API_KEY=cai_YOUR_API_KEY_HERE

Authentication

Centre AI supports two authentication methods:

API Key (Recommended for MCP)

API Keys start with cai_ and are passed in the header:

X-API-Key: cai_your_api_key_here

JWT Token

For web applications, use JWT tokens:

Authorization: Bearer eyJhbGciOiJIUzI1NiIs...

API Endpoints

Authentication

Memories

Projects & Tasks

Billing

MCP Tools Reference

Centre AI provides over 20 MCP tools. Here are the main categories:

Memory Management

Knowledge Graph

Code Indexing

Project Management

Web & Documentation

Conversations

Code Examples

Create Memory (cURL)

curl -X POST https://centre-ai.de/api/memories \
  -H "X-API-Key: cai_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "User prefers dark themes",
    "memory_type": "preference",
    "tags": ["ui", "settings"],
    "importance": 7
  }'

Semantic Search

curl -X GET "https://centre-ai.de/api/memories?query=theme+preferences&semantic=true" \
  -H "X-API-Key: cai_your_api_key"

Python Example

import requests

API_KEY = "cai_your_api_key"
BASE_URL = "https://centre-ai.de/api"

headers = {
    "X-API-Key": API_KEY,
    "Content-Type": "application/json"
}

# Create a memory
response = requests.post(
    f"{BASE_URL}/memories",
    headers=headers,
    json={
        "content": "User prefers Python over JavaScript",
        "memory_type": "preference",
        "tags": ["programming", "languages"],
        "importance": 8
    }
)

print(response.json())

JavaScript Example

const API_KEY = 'cai_your_api_key';
const BASE_URL = 'https://centre-ai.de/api';

async function createMemory(content, tags = []) {
  const response = await fetch(`${BASE_URL}/memories`, {
    method: 'POST',
    headers: {
      'X-API-Key': API_KEY,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      content,
      memory_type: 'general',
      tags,
      importance: 5
    })
  });

  return response.json();
}

// Usage
createMemory('Important project deadline: Dec 31')
  .then(console.log);

Troubleshooting

Connection Issues

Common Errors

Need Help?

Contact us at till.tmk@gmail.com

Documentation updated: December 2025