SQLite MCP Server

Reference MCP server for querying and analyzing SQLite databases through Claude and other MCP clients.

Free· Free / open source (MIT). No usage fees; runs locally against a SQLite file you control.MCP Servers

Best for

Developers and analysts who want an LLM to explore a local SQLite database through Claude Desktop or VS Code, and MCP server authors looking for a small, readable reference implementation to fork.

Skip if

Teams that need a maintained, hardened database connector, non-SQLite engines, multi-user access controls, or any production system where an archived upstream is a non-starter.

The SQLite MCP Server is a reference Model Context Protocol server originally maintained by the MCP core team that turns any local SQLite database into a first-class tool surface for Claude Desktop, VS Code and other MCP-compatible clients. It exposes a small, focused toolbelt (`read_query`, `write_query`, `create_table`, `list_tables`, `describe_table`) plus an `append_insight` tool and a `memo://insights` resource that accumulates business findings across a conversation, so the model can build up a running analysis memo instead of restating context each turn. A bundled `mcp-demo` prompt walks the model through a guided business-intelligence workflow against a topic you supply.

It is aimed at developers, analysts and prompt engineers who want to let an LLM explore a real relational dataset without hand-rolling a bespoke integration — think ad-hoc data exploration, prototyping agentic BI flows, teaching an LLM to write SQL against a schema it can introspect, or wiring a local SQLite cache into a larger agent stack. Deployment is deliberately trivial: `uv run mcp-server-sqlite --db-path ~/test.db`, a Docker image with a mounted volume, or the one-click VS Code install buttons in the README. Configuration snippets for Claude Desktop and VS Code are included.

Important caveat: the repository was archived on 29 May 2025 and moved to `servers-archived`, alongside other early reference servers that the MCP team retired in favour of community-maintained implementations. The code still works and remains under MIT license, but it will no longer receive fixes, security patches or feature work from the upstream maintainers, so treat it as a stable reference rather than a production dependency.

Editor's take

A tidy, honest reference server — the six-tool surface plus the insights-memo pattern is genuinely instructive, and it is still the shortest path from a .db file to a Claude conversation that can actually query it. Just treat the archived status seriously: fine for local exploration and as a template, not something I would pin a production agent to without forking.

— The AI Tool Bible editorial team

Pros

✅ Zero-config local database access for any MCP client — point it at a .db file and Claude can query, schema-introspect and write immediately.
✅ Clean, minimal tool surface (six tools) that maps cleanly to how an LLM actually reasons about a database.
✅ Novel `append_insight` + `memo://insights` pattern gives the model a persistent scratchpad for multi-turn analysis.
✅ MIT-licensed and open source, so it doubles as a canonical example for building your own MCP server.
✅ Multiple install paths — uv, Docker, VS Code one-click — cover most developer setups.
✅ No API keys, no cloud, no per-query cost; everything runs on your machine against a file you own.

Cons

⚠️ Repository was archived on 29 May 2025 — no more upstream fixes, security patches or new features.
⚠️ Exposes `write_query` and `create_table` to the model, so a careless prompt can mutate or drop data; there is no built-in read-only mode or row-level safety.
⚠️ SQLite-only — no Postgres, MySQL, DuckDB or cloud-warehouse support; you need a different MCP server for those.
⚠️ No authentication, quota or audit layer; intended for local single-user use, not shared/multi-tenant deployments.
⚠️ Insight-memo state lives in the running server process, so it does not survive restarts or multiple concurrent clients cleanly.

Use cases

Ad-hoc SQL exploration of a local SQLite database from Claude DesktopPrototyping agentic business-intelligence workflowsTeaching an LLM to write SQL against an introspected schemaBuilding a running insights memo across a multi-turn analysis sessionReference implementation for authoring a custom MCP serverWiring a local SQLite cache into a larger MCP-based agent stackQuick schema documentation via `list_tables` and `describe_table`Lightweight data prep and table creation inside a chat session