GitHub - dosco/graphjin: GraphJin - A Compiler to Connect AI to Your Databases

GraphJin - A Compiler to Connect AI to Your Databases

Point GraphJin at any database and AI assistants can query it instantly. Auto-discovers your schema, understands relationships, compiles to optimized SQL. No configuration required.

Works with PostgreSQL, MySQL, MongoDB, SQLite, Oracle, MSSQL - and models from Claude/GPT-4 to local 7B models.

Installation

npm (all platforms)

macOS (Homebrew)

brew install dosco/graphjin/graphjin

Windows (Scoop)

scoop bucket add graphjin https://github.com/dosco/graphjin-scoop
scoop install graphjin

Linux

Download .deb/.rpm from releases

Docker

docker pull dosco/graphjin

Try It Now

This is a quick way to try out GraphJin we'll use the --demo command which automatically starts a database using docker and loads it with demo data.

Download the source which contains the webshop demo

git clone https://github.com/dosco/graphjin
cd graphjin

Now launch the Graphjin service that you installed using the install options above

graphjin serve --demo --path examples/webshop

You'll see output like this:

GraphJin started
───────────────────────
  Web UI:      http://localhost:8080/
  GraphQL:     http://localhost:8080/api/v1/graphql
  REST API:    http://localhost:8080/api/v1/rest/
  MCP:         http://localhost:8080/api/v1/mcp

Claude Desktop Configuration
────────────────────────────
Add to claude_desktop_config.json:

  {
    "mcpServers": {
      "Webshop Development": {
        "command": "/path/to/graphjin",
        "args": ["mcp", "--server", "http://localhost:8080"]
      }
    }
  }

Copy the JSON config shown and add it to your Claude Desktop config file (see below for file location). You can also click File > Settings > Developer to get to it in Claude Desktop. You will also need to Restart Claude Desktop

Getting started

To use GraphJin with your own databases you have to first create a new GraphJin app, then configure it using its config files and then launch GraphJin.

Step 1: Create New GraphJin App

Step 2: Start the GraphJin Service

graphjin serve --path ./my-app

Step 3: Add to Claude Desktop config file

Copy paste the Claude Desktop Config provided by graphjin serve into the Claude Desktop MCP config file. How to do this has been defined clearly above in the Try it Now section.

Step 4: Restart Claude Desktop

Step 5: Ask Claude questions like:

• "What tables are in the database?"
• "Show me all products under $50"
• "List customers and their purchases"
• "What's the total revenue by product?"
• "Find products with 'wireless' in the name"
• "Add a new product called 'USB-C Cable' for $19.99"

How It Works

1. Connects to database - Reads your schema automatically
2. Discovers relationships - Foreign keys become navigable joins
3. Exposes MCP tools - Teach any LLM the query syntax
4. Compiles to SQL - Every request becomes a single optimized query

No resolvers. No ORM. No N+1 queries. Just point and query.

What AI Can Do

Simple queries with filters:

{ products(where: { price: { gt: 50 } }, limit: 10) { id name price } }

Nested relationships:

{
  orders(limit: 5) {
    id total
    customer { name email }
    items { quantity product { name category { name } } }
  }
}

Aggregations:

{ products { count_id sum_price avg_price } }

Mutations:

mutation {
  products(insert: { name: "New Product", price: 29.99 }) { id }
}

Spatial queries:

{
  stores(where: { location: { st_dwithin: { point: [-122.4, 37.7], distance: 1000 } } }) {
    name address
  }
}

Real-time Subscriptions

Get live updates when your data changes. GraphJin handles thousands of concurrent subscribers with a single database query - not one per subscriber.

subscription {
  orders(where: { user_id: { eq: $user_id } }) {
    id total status
    items { product { name } }
  }
}

Why it's efficient:

• Traditional approach: 1,000 subscribers = 1,000 database queries
• GraphJin: 1,000 subscribers = 1 optimized batch query
• Automatic change detection - updates only sent when data actually changes
• Built-in cursor pagination for feeds and infinite scroll

Works from Node.js, Go, or any WebSocket client.

MCP Tools

GraphJin exposes several tools that guide AI models to write valid queries. Key tools: list_tables and describe_table for schema discovery, get_query_syntax for learning the DSL, execute_graphql for running queries, and execute_saved_query for production-approved queries. Prompts like write_query and fix_query_error help models construct and debug queries.

Database Support

Also works with AWS Aurora/RDS, Google Cloud SQL, and YugabyteDB.

Production Security

Query allow-lists - In production, only saved queries can run. AI models call execute_saved_query with pre-approved queries. No arbitrary SQL injection possible.

Role-based access - Different roles see different data:

roles:
  user:
    tables:
      - name: orders
        query:
          filters: ["{ user_id: { eq: $user_id } }"]

JWT authentication - Supports Auth0, Firebase, JWKS endpoints.

Response caching - Redis with in-memory fallback. Automatic cache invalidation.

Also a GraphQL API

GraphJin works as a traditional API too - use it from Go or as a standalone service.

Go

go get github.com/dosco/graphjin/core/v3

db, _ := sql.Open("pgx", "postgres://localhost/myapp")
gj, _ := core.NewGraphJin(nil, db)
res, _ := gj.GraphQL(ctx, `{ users { id email } }`, nil, nil)

Standalone Service

brew install dosco/graphjin/graphjin  # Mac
graphjin new myapp && cd myapp
graphjin serve

Built-in web UI at http://localhost:8080 for query development.

Documentation

• Configuration Reference
• Feature Reference
• Go Examples

Get in Touch

Twitter @dosco | Discord

License

Apache Public License 2.0