ZincBind

ZincBind

MCP server exposing ZincBindDB GraphQL API for zinc binding sites and related data.

0
Stars
0
Forks
0
Releases

Overview

ZincBind MCP Server is a Model Context Protocol (MCP) server that provides programmatic access to the ZincBindDB GraphQL API for querying zinc binding site structures and protein data. The server runs on Cloudflare Workers, enabling scalable, serverless access for AI assistants and automated workflows to explore zinc binding sites, PDB entries with zinc coordination, multi-metal sites and chain clusters, as well as liganding residues and binding geometries. It exposes one main MCP tool, zincbind_db_graphql_query, to execute GraphQL queries against the ZincBindDB API, including support for schema introspection and common queries such as listing binding sites and retrieving related PDB structures. The repository includes development and deployment scripts (dev, deploy, format, lint:fix, cf-typegen) and a core MCP agent class (ZincBindMCP) with a custom GraphQL client for executing queries. This MCP server is designed for researchers and AI developers who want to integrate ZincBindDB data into training pipelines or conversational assistants, with setup guided for local development and Cloudflare deployment. It is MIT licensed with citation requirements.

Details

Owner
QuentinCody
Language
TypeScript
License
Other
Updated
2025-12-07

Features

ZincBindDB GraphQL access

Programmatic access to query zinc binding site structures, PDB entries, and related ZincBindDB data via GraphQL.

Cloudflare Workers runtime

Runs the MCP server on Cloudflare Workers for scalable, serverless deployment.

ZincBindDB data coverage

Includes zinc binding sites, PDB entries with zinc coordination, multi-metal sites, chain clusters, and ligand residues and binding geometries.

zincbind_db_graphql_query tool

Core MCP tool to execute GraphQL queries against the ZincBindDB API.

GraphQL introspection support

Supports schema introspection to discover available queries and fields.

Local development workflow

Hot-reload development server with npm run dev for rapid iteration.

Deployment workflow

Deploy to Cloudflare Workers with npm run deploy.

Developer tooling

Code formatting and linting via Biome; type generation with cf-typegen.

Audience

AI developersIntegrate ZincBindDB queries into AI assistants via MCP server for zinc-binding data.
ResearchersAccess comprehensive zinc binding site data for metalloproteomics analysis and structure prediction.
Cloudflare engineersManage deployment, scaling, and CI/CD for the MCP server on Cloudflare Workers.

Tags

ZincBindZincBindDBMCPGraphQLCloudflare WorkersPDBzinc coordinationmulti-metal sitesliganding residuesbinding geometries