Skip to main content
Glama

Chemist Model Context Protocol (MCP) Server

TypeScript MCP License: MIT

A Model Context Protocol (MCP) server providing deep LLM-agent tool, resource, and prompt integrations with Chemist — an open-source computational research platform and quantitative analytical laboratory for comparative chemistry, molecular structures, and functional group manifolds.


Overview

The chemist-mcp server enables AI agents (Claude Code, Cursor, Gemini CLI, Antigravity) to query, manipulate, cluster, and synthesize chemistry domain models directly over standard input/output (stdio):

  • 51 Registered MCP Tools across 8 functional modules (Search, Chemical Classes, Molecules, Functional Groups, Clustering, Synthesis Reports, Background Jobs, and System Diagnostics).

  • 4 Registered MCP Resources exposing read-only live platform status, catalogs, cluster snapshots, and markdown monographs.

  • 2 MCP Prompt Templates for comparative chemistry studies and cluster synthesis monographs.

  • 1024-dimensional semantic manifold querying via PostgreSQL pgvector HNSW indexes.


Related MCP server: Mol-MCP

Architectural Mapping

Tier / Domain

Model

Table Name

MCP Module / Tools

Tier 1

Chemical Superfamilies

chemical_classes

list_chemical_classes, get_chemical_class, create_chemical_class, update_chemical_class, delete_chemical_class, upsert_chemical_classes_batch, populate_chemical_classes, populate_chemical_class_molecules

Tier 2

Molecules / Compounds

molecules

list_molecules, get_molecule, create_molecule, update_molecule, delete_molecule, upsert_molecules_batch, generate_molecule_embedding, populate_molecule_functional_groups, populate_molecules_bulk

Tier 3

Functional Groups / Reactive Motifs

functional_groups

list_functional_groups, get_functional_group, create_functional_group, update_functional_group, delete_functional_group, upsert_functional_groups_batch, generate_functional_group_embedding, populate_functional_groups_bulk

Clustering

15-Level Agglomerative

cluster_snapshots, cluster_groups

compute_clusters, create_cluster_snapshot, list_cluster_snapshots, get_cluster_snapshot, delete_cluster_snapshot, list_cluster_groups, get_cluster_group

Synthesis

Multi-Stage Map-Reduce

cluster_reports, cluster_report_summaries

list_cluster_reports, get_cluster_report, create_cluster_report, delete_cluster_report, generate_deep_synthesis, list_report_summaries, get_report_summary, list_report_sub_reports, get_report_sub_report

Global Search

Cross-Catalog pgvector

Multi-table

search, find_counterparts

Platform

Jobs & System Info

Multi-service

list_jobs, get_job_status, wait_for_job, get_platform_status, get_settings, update_context_window, trigger_bulk_embeddings, get_openapi_spec


Installation & Setup

Prerequisites

  • Node.js 20+

  • Running instance of Chemist (http://localhost:3000)

Build

git clone https://github.com/leouofa/chemist-mcp.git
cd chemist-mcp
npm install
npm run build

Configuration

Environment Variables

Configure via .env file or export in your shell:

# Base URL for the Chemist Rails API (v1)
CHEMIST_API_URL=http://localhost:3000/api/v1

# Optional API key if authentication is enabled
# CHEMIST_API_KEY=

# HTTP request timeout in milliseconds (default: 60000)
CHEMIST_TIMEOUT_MS=60000

CLI Arguments

chemist-mcp [options]

Options:
  -u, --url <url>        Chemist API base URL (default: http://localhost:3000/api/v1)
  -k, --api-key <key>    Optional API key / Bearer token
  -t, --timeout <ms>     Request timeout in milliseconds (default: 60000)
  -h, --help             Display help message

MCP Client Configuration

Claude Desktop / Antigravity / Cursor Configuration

Add to your client configuration (e.g. claude_desktop_config.json or MCP settings):

{
  "mcpServers": {
    "chemist": {
      "command": "node",
      "args": ["/path/to/chemist-mcp/dist/index.js"],
      "env": {
        "CHEMIST_API_URL": "http://localhost:3000/api/v1"
      }
    }
  }
}

MCP Tools Reference (51 Tools)

Search (2 tools)

  • search: Cross-catalog hybrid semantic (1024-d cosine similarity) and keyword search across molecules and functional groups.

  • find_counterparts: Find nearest bioisosteric analogs or reactive motifs using vector cosine distance.

Chemical Classes (8 tools)

  • list_chemical_classes: Query superfamilies with molecule counts and filters.

  • get_chemical_class: Detailed chemical class view with associated molecules.

  • create_chemical_class: Register a new chemical superfamily.

  • update_chemical_class: Update name or description.

  • delete_chemical_class: Remove chemical class and cascaded children.

  • upsert_chemical_classes_batch: Bulk insert/update chemical classes.

  • populate_chemical_classes: Trigger LLM discovery agent for superfamilies.

  • populate_chemical_class_molecules: Discover molecules for a specific chemical class.

Molecules (9 tools)

  • list_molecules: List molecules with filter by chemical class, themes, and embedding status.

  • get_molecule: Detailed molecule view with child functional groups and embeddings.

  • create_molecule: Register a new compound.

  • update_molecule: Update name, description, themes, or parent chemical class.

  • delete_molecule: Delete molecule and child functional groups.

  • upsert_molecules_batch: Bulk upsert molecules under a chemical class.

  • generate_molecule_embedding: Generate or refresh 1024-d vector embedding.

  • populate_molecule_functional_groups: Discover reactive motifs for a molecule.

  • populate_molecules_bulk: Bulk discover molecules across chemical classes.

Functional Groups (8 tools)

  • list_functional_groups: List reactive motifs with thematic filtering.

  • get_functional_group: Detailed functional group view.

  • create_functional_group: Register a new functional group under a molecule.

  • update_functional_group: Update name, description, or themes.

  • delete_functional_group: Remove functional group record.

  • upsert_functional_groups_batch: Bulk upsert functional groups for a molecule.

  • generate_functional_group_embedding: Generate or refresh 1024-d vector embedding.

  • populate_functional_groups_bulk: Bulk discover functional groups.

Clusters & Snapshots (7 tools)

  • compute_clusters: Preview live connected-components clusters across 15 levels.

  • create_cluster_snapshot: Freeze cluster configuration into database snapshot.

  • list_cluster_snapshots: List frozen snapshots.

  • get_cluster_snapshot: Retrieve snapshot with groups and member entities.

  • delete_cluster_snapshot: Delete cluster snapshot.

  • list_cluster_groups: List archetype groups for a snapshot.

  • get_cluster_group: Detailed cluster group inspection.

Synthesis Reports (9 tools)

  • list_cluster_reports: List multi-phase synthesis monographs.

  • get_cluster_report: Retrieve full markdown monograph and progress state.

  • create_cluster_report: Initiate 2-Phase Map-Reduce synthesis.

  • delete_cluster_report: Remove report and associated summaries.

  • generate_deep_synthesis: Initiate 3-Phase Hierarchical Deep Synthesis monograph.

  • list_report_summaries: List Phase 1 per-cluster analyses.

  • get_report_summary: Retrieve single cluster analysis.

  • list_report_sub_reports: List Phase 2 batch syntheses.

  • get_report_sub_report: Retrieve single batch synthesis.

Background Jobs (3 tools)

  • list_jobs: List recent background Sidekiq/ActiveJob tasks.

  • get_job_status: Poll job status and progress metadata.

  • wait_for_job: Wait synchronously for job completion with timeout.

Platform & Settings (5 tools)

  • get_platform_status: Health check, entity counts, active model parameters.

  • get_settings: Vector embedding metrics and LLM context window presets.

  • update_context_window: Update context token budget.

  • trigger_bulk_embeddings: Enqueue bulk embedding calculation.

  • get_openapi_spec: Retrieve complete raw OpenAPI specification.


MCP Resources Reference (4 Resources)

  • chemist://status: Real-time JSON health check, model configuration, and database item counters.

  • chemist://chemical-classes: JSON catalog of top registered chemical classes.

  • chemist://snapshots/{id}: Detailed JSON payload for a frozen cluster snapshot with member groups.

  • chemist://reports/{id}: Formatted Markdown synthesis monograph.


MCP Prompts Reference (2 Prompts)

  • comparative_chemical_analysis: Structured workflow comparing two molecules or functional groups through stereoelectronic, biosynthetic, and vector cosine distance lenses.

  • chemical_cluster_monograph: Guided investigation of an agglomerative cluster snapshot producing an executive monograph with citations to specific cluster groups.


License

MIT License. Copyright (c) 2026 Leonid Medovyy.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server for chemistry-focused tools, enabling LLM agents to perform molecule parsing, format conversion, property lookup, and other chemistry operations with explainable responses.
      Apache 2.0
    • F
      license
      B
      quality
      D
      maintenance
      Provides computational chemistry tools for LLMs, enabling molecular operations like SMILES processing and geometry manipulation via a modular agent and tool system.
      9
      -
    • A
      license
      C
      quality
      D
      maintenance
      Enables chemical reaction predictions (retrosynthesis, product, reagents), molecule validation, and property calculation through the Model Context Protocol, integrating with AI assistants.
      5
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Provides in-process RDKit cheminformatics operations — descriptors, fingerprints, similarity and substructure search, reaction enumeration, standardization, and 3D conformer generation — for AI agents.
      -