SQLite-Anet-MCP Server
About
SQLite-Anet-MCP Server A blazing-fast, Rust-powered SQLite server for AI agents—speak JSON-RPC, store insights, and manage your database like a pro.
Details
- Author
- marekkucak
- Downloads
- 324
- Categories
- Database, AI
Jump to
- Create and manage SQLite database tables
- Execute SELECT queries for data retrieval
- Execute INSERT, UPDATE, and DELETE queries for data manipulation
- Describe table schemas and list available tables
- Save and synthesize business insights from data
- NATS transport layer for message passing
Clone the repository, configure a .env file with NATS_URL, MCP_SUBJECT, SQLITE_DB_PATH, and RUST_LOG, ensure a NATS server is running, then run cargo run. Test using cargo run --example test_client or cargo run --example chinook_test after downloading the Chinook database.
SQLite-Anet-MCP Server
A Rust implementation of the Model Control Protocol (MCP) server that provides SQLite database capabilities via a standardized protocol. This server enables AI agents to create, manage, and query SQLite databases directly.
This project is based on the Model Context Protocol SQLite Server reference implementation.
---
Features
- 🗃️ Create and manage SQLite database tables
- 🔍 Execute SELECT queries for data retrieval
- ✏️ Execute INSERT, UPDATE, and DELETE queries for data manipulation
- 📊 Describe table schemas and list available tables
- 📝 Save and synthesize business insights from data
- 🔄 NATS transport layer for message passing
- 🛠️ JSON-RPC 2.0 compatible API
- ⚡ Asynchronous request handling with Tokio
---
Requirements
- Rust 1.70+
- NATS server running locally or accessible via network
- SQLite (included as a Rust dependency)
---
Installation
Clone the repository and build the server:
git clone https://github.com/yourusername/sqlite-anet-mcp.git
cd sqlite-anet-mcp
Configure your environment in a .env file:
NATS_URL=nats://localhost:4222
MCP_SUBJECT=mcp.requests
SQLITE_DB_PATH=./data/sqlite.db
RUST_LOG=debug
---
Getting Started
Running the Server
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.



