beingdb-mcp
Mentions that the nginx reverse proxy can be placed behind Apache for TLS and further access control.
Provides integration with Docker via docker-compose to orchestrate the BeingDB HTTP API, the beingdb-mcp server, and an nginx reverse proxy.
Provides integration with NGINX as a reverse proxy, adding rate limiting (10 req/s) and serving as a front-end for both the BeingDB REST API and the MCP endpoint.
Notes that the deployment (with Docker Compose) is suitable for headless hosts like a Raspberry Pi.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
beingdb-mcp
beingdb-mcp lets MCP-capable clients such as VS Code Copilot query a
BeingDB database. The connected LLM uses its
tools to discover the database's schema, write queries, check them and run them.
beingdb-mcp contains no LLM itself; it passes requests to BeingDB and returns
BeingDB's answers.
For development and testing, BeingDB and the MCP server both run locally on your machine. You only need a deployed server when you want to make the service available to other machines or users.
Quick start
You need Node.js 22+ and a BeingDB checkout
(with its OCaml tools, opam and dune) next to this repository:
git/
beingdb/
beingdb-mcp/In one terminal:
cd beingdb-mcp
npm install
npm run beingdb:localnpm run beingdb:local builds the BeingDB checkout next to this one and serves the
bundled demo data on 127.0.0.1:8080.
In a second terminal:
npm run devThis starts the MCP server at http://127.0.0.1:3000/mcp. Stop either with Ctrl-C.
VS Code
↓
beingdb-mcp
↓
BeingDB
↓
local dataEverything in this setup runs on your own machine.
Related MCP server: NL-to-SQL MCP
Connect VS Code
This repository includes a workspace MCP configuration, .vscode/mcp.json, so you
don't need to change MCP settings used by unrelated projects:
{
"servers": {
"beingdb": {
"type": "http",
"url": "http://127.0.0.1:3000/mcp"
}
}
}Open the project in VS Code, start the beingdb MCP server from that file (or with
MCP: List Servers), and use Copilot in agent mode.
If you previously configured a global BeingDB MCP server, disable or remove it when
using this workspace configuration to avoid having two beingdb servers configured.
What it provides
beingdb_list_predicates: discover the schemabeingdb_resolve: find the stored values the names in a question refer tobeingdb_about: list everything about one valuebeingdb_diagnose_query: check a query and get diagnostics and proven repairsbeingdb_query: execute a querybeingdb_validate_query: validate a querybeingdb_explain_query: inspect its planbeingdb_get_facts: retrieve a predicate's factsbeingdb_status: check the connection
A typical session:
discover schema
↓
resolve the names in the question
↓
write query
↓
diagnose if needed
↓
executeSee docs/tools.md for details.
Tests
npm testThis runs the normal test suite locally and starts and stops its own BeingDB
instance; you don't need the Quick start servers running. npm run test:unit runs
only the unit tests, without BeingDB.
Using your own data
npm run beingdb:local -- --pack /path/to/pack_storeThis serves your own compiled BeingDB pack instead of the demo data.
Deployment
The local setup above is intended for development and testing.
beingdb-mcp can also be deployed to a server when you want other machines or users
to access it. The same MCP tools are used; the MCP server simply connects to a
deployed BeingDB instead of localhost:
VS Code / MCP client
↓
hosted beingdb-mcp
↓
hosted BeingDBSee docs/deployment.md.
Documentation
docs/tools.md: tools, outputs, schema metadata, diagnostics and repairs
docs/prompting.md: prompting an LLM to use BeingDB carefully, including a REWIND example
docs/development.md: local setup details, configuration, tests and troubleshooting
docs/deployment.md: hosting BeingDB and the MCP server
License
This server cannot be deployed
Maintenance
Related MCP Connectors
Hosted MCP server for live public-data APIs and Skills for AI agents.
Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.
AI Reasoning Cache & Consensus Layer with 11 MCP tools via Streamable HTTP.
MCP server exposing the Backtest360 engine API as tools for AI agents.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceExposes your local machine's filesystem, git, shell, network, databases, and system to any MCP-compatible LLM client over HTTP.7-
- FlicenseNot gradedqualityDmaintenanceMCP server enabling natural-language querying of SQLite databases via schema discovery, GraphRAG retrieval, and safely guarded read-only SQL execution.-
- AlicenseAqualityCmaintenanceExposes internal databases, HTTP APIs, and files to LLM agents over MCP, with auto-generated schemas, access controls, rate limiting, redaction, and audit logging.3MIT
- AlicenseNot gradedqualityCmaintenanceExposes remote, agent-facing MCP endpoints over Streamable HTTP for web search, webpage parsing, local knowledge-base retrieval and document creation, and delegation of tasks to configurable LLM providers. Also bundles an admin console with onboarding, provider/model configuration, API-key management, and call logging.MIT