MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
The open computational chemistry engine for drug discovery and materials science; MCP + REST + Workbench + Next.js dashboard, wrapping RDKit, GROMACS, AutoDock, OpenFold, Chai, Boltz, Gnina.
pypi NOASSERTION ★ 6 -
材料科学に特化した MCP サーバー群 / MCP servers for materials science
-
sci-host-mcpgithub.com/aceris-sola/sci-host-mcpupdated this year
Standalone MCP service for auditable materials-science literature research
-
scimcpgithub.com/Nour-elhaq/scimcpupdated this year
MCP server for computational materials science LAMMPS, DFT, and ML powered analysis
MIT ★ 1 -
Production-grade Science of Reading MCP server & Material Design 3 Web App for K-5 literacy analysis, 50-state Standards Satchel alignment, decodability checking, and FERPA-compliant PII anonymization.
-
MCP for the Materials Project
BSD-3-Clause ★ 1 -
mcp-materials-projectgithub.com/luffysolution-svg/mcp-materials-projectupdated this year
MCP Server for Materials Project database - query material properties, crystal structures, and phase diagrams with horizontal comparison Excel export
pypi MIT -
matroutergithub.com/matrouter/matrouterupdated this year
Local-first Python package for task-relative materials evidence routing with MCP integration.
pypi Apache-2.0 -
MCP server with 25 tools and a two phase research agent over Materials Project. Personal course project.
MIT -
Local MCP research toolkit for literature, Zotero, PDFs, citations, materials, and scientific figures
npm MIT ★ 8 -
A graph-style materials research framework backend that supports continuous from-prompt building.
-
npx launchers for 5 Huginn MCP servers — materials database, math semantics, pixel vision, capability containers, workflow blueprints
npm MIT -
mace-mcp-servergithub.com/zichengzhao/mace-mcp-serverupdated this year
MCP server for MACE machine learning interatomic potentials
pypi MIT -
LLM 调度 + 确定性 MCP 工具层的材料计算助手(pymatgen / ASE / VASP 输入解析与工作流)
MIT ★ 1 -
—
npm MIT -
Verified DFT workflow MCP toolchain: dual-engine (GPAW/QE), convergence gates that actually gate, runtime results-vs-Materials-Project reconciliation, full evidence chain.
MIT ★ 3 -
A Model Context Protocol server for ABACUS first-principles calculations
pypi MIT -
failure-analysis-agent-mcpupdated this year
Evidence-gated failure analysis MCP connector for austenitic stainless steel cases
npm MIT -
paper-hermesgithub.com/elomokm/paper-hermesupdated this year
MCP Server de veille scientifique — arXiv/ChemRxiv → DeepSeek → Brain Obsidian → Telegram. 8 tools, cron 08:00
-
POLARIS FastAPI backend: LLM research agents (hypothesis, analysis, ML, curve fitting, literature), experiment watcher, MCP orchestrator, and phase-mapper pipeline.
-
Agent-driven single-crystal X-ray structure solution and refinement workbench
NOASSERTION ★ 2 -
Free high-entropy alloy (HEA) and high-entropy oxide calculator: mixing entropy, atomic-size mismatch, VEC, Miedema enthalpy, and the canonical empirical phase-prediction rules. Python library, browser app, desktop app, and MCP server.
pypi MIT ★ 3 -
MCP server to check VASP inputs before you submit and read the results after: convergence, energies, forces, band gaps per spin, crash causes. Read-only.
MIT -
LLM agent orchestration for scientific workflows. Python 3.12, LiteLLM, BoTorch, ASE, Streamlit UI, MCP server and client.
-
dft-autopilot-mcpgithub.com/The66user/dft-autopilot-mcpupdated this year
BYO-HPC DFT computational chemistry (Quantum ESPRESSO / VASP / Gaussian) orchestration — an MCP server + Claude plugin that runs on your own HPC (SCNet / Slurm / local).
Apache-2.0 ★ 3 -
Untiring, token-efficient research collaborator: reads papers and patents, cites the paragraph, writes the paper, designs and simulates what it proposes.
pypi GPL-3.0 ★ 11 -
物理约束的能带结构 MCP 服务器:让 AI 读取论文/实验能带图与数值能带,提供检索、匹配与可信拒识的开源工具。A physics-aware band-structure MCP server for AI agents (Docker, stdio/HTTP, MIT). Built with AI-agent assistance by a mechanical-engineering student as a DFT-learning project.
MIT -
AutoPhaseAnalysisgithub.com/jojusmathew/AutoPhaseAnalysisupdated this year
Automated XRD phase identification and Rietveld refinement — a dara tree search explores phase combinations and BGMN-refines each one, no manual peak matching. Ranked solutions with interactive Plotly charts, phase-weight tables, exportable refinement files, and an MCP server for headless AI-agent control.
NOASSERTION ★ 1 -
AI-agent-native low cycle fatigue strain-life analysis. Python library plus MCP server: reduce strain-controlled test data, fit Basquin, Coffin-Manson, and Ramberg-Osgood models, predict life with mean-stress corrections, and save results for recall.
pypi MIT -
GSAS-II powder-XRD Rietveld refinement for AI agents: 4 skills (single-phase, multiphase QPA, publication plotting, automated workflow) + an MCP server exposing 19 GSAS-II tools. Refinement / phase judgement / plotting are strictly separated; every reported number carries covariance esd and file hashes.
★ 2 -
Crystal and magnetic structure refinement in the browser: Rietveld, single-crystal F², magnetic structures and PDF/mPDF on one TypeScript engine, checked against GSAS-II, FullProf, SHELXL and PDFfit2. With an AI Copilot (local or cloud models, 60 tools) and a 41-tool MCP server. Nothing to install.
AGPL-3.0 -
📚 Simplify LLM access to the M.U.S.T. campus system, enabling smooth interactions with schedules, assignments, and course materials.
MIT ★ 2 -
Work in progress, addressable: a self-hosted workspace where humans and AI agents work the same material while it is still moving. No API keys, no metered inference.
npm Apache-2.0 ★ 1