pubchem-mcp-server

pubchem-mcp-server

3.2

If you are the rightful owner of pubchem-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The PubChem MCP Server is a Model Context Protocol server that provides seamless integration with PubChem's chemical information database, enabling AI agents and scientific tools to access and analyze chemical data efficiently.

PubChem MCP Server

Empower your AI agents and scientific tools with seamless PubChem integration!

An MCP (Model Context Protocol) server providing comprehensive access to PubChem's vast chemical information database. Enables LLMs and AI agents to search, retrieve, and analyze chemical compounds, substances, and bioassays through the PubChem PUG REST API.

Built on the cyanheads/mcp-ts-template, this server follows a modular architecture with robust error handling, logging, and security features.

šŸš€ Core Capabilities: PubChem Tools šŸ› ļø

This server equips your AI with specialized tools to interact with PubChem:

Tool NameDescriptionKey Features
Searches for PubChem Compound IDs (CIDs) using a common chemical identifier.- Search by name, smiles, or inchikey.
- The primary entry point for most compound-based workflows.
Fetches a list of specified physicochemical properties for one or more CIDs.- Retrieve properties like MolecularWeight, XLogP, IUPACName, etc.
- Essential for gathering detailed chemical data in bulk.
Fetches a 2D image of a compound's structure for a given CID.- Returns the raw image data as a binary blob.
- Supports small (100x100) and large (300x300) image sizes.
Performs a structural search using a SMILES string or a CID as the query.- Supports substructure, superstructure, and identity search types.
- Essential for finding structurally related compounds.
Finds compounds with a similar 2D structure to a query compound.- Based on a Tanimoto similarity score.
- Search by smiles or cid.
- Configurable threshold and maxRecords.
Finds PubChem Compound IDs (CIDs) that match a given molecular formula.- Supports exact matches and formulas with additional elements.
- Configurable maxRecords.
Retrieves details for a given PubChem Substance ID (SID).- Fetches synonyms, source, dates, and related CIDs.
Fetches a detailed summary for a specific PubChem BioAssay ID (AID).- Includes name, description, source, and statistics.
Finds PubChem BioAssay IDs (AIDs) associated with a specific biological target.- Search by genesymbol or proteinname.
Fetches external cross-references (XRefs) for a given CID.- Retrieve RegistryID, PubMedID, PatentID, etc.
- Supports pagination for large result sets.

Table of Contents

| Overview | Features | Installation |

| Configuration | Project Structure |

| Tools | Development | License |

Overview

The PubChem MCP Server acts as a bridge, allowing applications (MCP Clients) that understand the Model Context Protocol (MCP) – like advanced AI assistants (LLMs), IDE extensions, or custom research tools – to interact directly and efficiently with PubChem's vast chemical database.

Instead of complex API integration or manual searches, your tools can leverage this server to:

  • Automate chemical research: Search for compounds, fetch detailed properties, find similar structures, and analyze bioassay results programmatically.
  • Gain chemical insights: Access detailed compound data, substance information, and bioassay metadata without leaving the host application.
  • Integrate PubChem into AI-driven science: Enable LLMs to conduct chemical research, analyze structure-activity relationships, and support evidence-based discovery.

Built on the robust mcp-ts-template, this server provides a standardized, secure, and efficient way to expose PubChem functionality via the MCP standard. It achieves this by integrating with the PubChem PUG REST API, ensuring compliance with rate limits and providing comprehensive error handling.

Developer Note: This repository includes a file that serves as a developer cheat sheet for your LLM coding agent with quick reference for the codebase patterns, file locations, and code snippets.

Features

Core Utilities

Leverages the robust utilities provided by the mcp-ts-template:

  • Logging: Structured, configurable logging (file rotation, stdout JSON, MCP notifications) with sensitive data redaction.
  • Error Handling: Centralized error processing, standardized error types (McpError), and automatic logging.
  • Configuration: Environment variable loading (dotenv) with comprehensive validation.
  • Input Validation/Sanitization: Uses zod for schema validation and custom sanitization logic.
  • Request Context: Tracking and correlation of operations via unique request IDs using AsyncLocalStorage.
  • Type Safety: Strong typing enforced by TypeScript and Zod schemas.
  • HTTP Transport: High-performance HTTP server using Hono, featuring session management with garbage collection and CORS support.
  • Authentication: Robust authentication layer supporting JWT and OAuth 2.1, with fine-grained scope enforcement.
  • Deployment: Multi-stage Dockerfile for creating small, secure production images with native dependency support.

PubChem Integration

  • PubChem PUG REST Integration: Comprehensive access to the PubChem API via a dedicated, rate-limited client.
  • Advanced Search Capabilities: Search by identifier, structure, similarity, and molecular formula.
  • Full Compound Data: Retrieve complete compound properties, including physicochemical data, names, and identifiers.
  • Substance and Assay Information: Fetch detailed records for substances (SIDs) and bioassays (AIDs).
  • Cross-Referencing: Find links to other databases like PubMed, patent registries, and gene databases.
  • Image Generation: Directly fetch 2D structural images of compounds.

Installation

Prerequisites

Install via npm (recommended)

npm install @cyanheads/pubchem-mcp-server

Alternatively Install from Source

  1. Clone the repository:

    git clone https://github.com/cyanheads/pubchem-mcp-server.git
    cd pubchem-mcp-server
    
  2. Install dependencies:

    npm install
    
  3. Build the project:

    npm run build
    *or npm run rebuild*
    

Configuration

Environment Variables

Configure the server using environment variables. These environmental variables are set within your MCP client config/settings (e.g. cline_mcp_settings.json for Cline).

VariableDescriptionDefault
MCP_TRANSPORT_TYPETransport mechanism: stdio or http.stdio
MCP_HTTP_PORTPort for the HTTP server (if MCP_TRANSPORT_TYPE=http).3010
MCP_HTTP_HOSTHost address for the HTTP server (if MCP_TRANSPORT_TYPE=http).127.0.0.1
MCP_ALLOWED_ORIGINSComma-separated list of allowed origins for CORS (if MCP_TRANSPORT_TYPE=http).(none)
MCP_LOG_LEVELLogging level (debug, info, notice, warning, error, crit, alert, emerg).info
LOG_OUTPUT_MODELogging output mode: file or stdout.file
MCP_AUTH_MODEAuthentication mode for HTTP: jwt or oauth.jwt
MCP_AUTH_SECRET_KEYRequired for jwt auth. Minimum 32-character secret key for JWT authentication.(none)
LOGS_DIRDirectory for log file storage (if LOG_OUTPUT_MODE=file).logs/

Note: The PubChem API does not require an API key for basic use, so no key is needed in the environment configuration.

MCP Client Settings

Add the following to your MCP client's configuration file (e.g., cline_mcp_settings.json). This configuration uses npx to run the server, which will automatically install the package if not already present:

{
  "mcpServers": {
    "pubchem-mcp-server": {
      "command": "npx",
      "args": ["@cyanheads/pubchem-mcp-server"],
      "env": {},
      "disabled": false,
      "autoApprove": []
    }
  }
}

Project Structure

The codebase follows a modular structure within the src/ directory:

src/
ā”œā”€ā”€ index.ts              # Entry point: Initializes and starts the server
ā”œā”€ā”€ config/               # Configuration loading (env vars, package info)
│   └── index.ts
ā”œā”€ā”€ mcp-server/           # Core MCP server logic and capability registration
│   ā”œā”€ā”€ server.ts         # Server setup, capability registration
│   ā”œā”€ā”€ transports/       # Transport handling (stdio, http)
│   └── tools/            # MCP Tool implementations (subdirs per tool)
ā”œā”€ā”€ services/             # External service integrations
│   └── pubchem/          # PubChem API client
ā”œā”€ā”€ types-global/         # Shared TypeScript type definitions
└── utils/                # Common utility functions (logger, error handler, etc.)

For a detailed file tree, run npm run tree or see .

Tools

The PubChem MCP Server provides a comprehensive suite of tools for chemical information retrieval, callable via the Model Context Protocol.

Tool NameDescriptionKey Arguments
pubchem_search_compound_by_identifierSearches for CIDs using an identifier (name, SMILES, InChIKey).identifierType, identifier
pubchem_fetch_compound_propertiesFetches physicochemical properties for one or more CIDs.cids, properties
pubchem_get_compound_imageFetches a 2D structural image for a given CID.cid, size?
pubchem_search_compounds_by_structurePerforms a structural search (substructure, superstructure, identity).searchType, query, queryType, maxRecords?
pubchem_search_compounds_by_similarityFinds compounds with a similar 2D structure to a query.query, queryType, threshold?, maxRecords?
pubchem_search_compounds_by_formulaFinds CIDs that match a given molecular formula.formula, allowOtherElements?, maxRecords?
pubchem_fetch_substance_detailsRetrieves details for a given Substance ID (SID).sid
pubchem_fetch_assay_summaryFetches a summary for a specific BioAssay ID (AID).aid
pubchem_search_assays_by_targetFinds BioAssay IDs (AIDs) associated with a biological target.targetType, targetQuery
pubchem_fetch_compound_xrefsFetches external cross-references for a given CID.cid, xrefTypes, page?, pageSize?

Note: All tools support comprehensive error handling and return structured JSON responses.

Development

Build and Test

# Build the project (compile TS to JS in dist/ and make executable)
npm run build

# Test the server locally using the MCP inspector tool (stdio transport)
npm run inspector

# Test the server locally using the MCP inspector tool (http transport)
npm run inspector:http

# Clean build artifacts
npm run clean

# Generate a file tree representation for documentation
npm run tree

# Clean build artifacts and then rebuild the project
npm run rebuild

# Format code with Prettier
npm run format

# Start the server using stdio (default)
npm start
# Or explicitly:
npm run start:stdio

# Start the server using HTTP transport
npm run start:http

License

This project is licensed under the Apache License 2.0 - see the file for details.


Built with the Model Context Protocol