What can you do with this server?

* LAMMPS Simulation Setup: Generate complete input scripts for potentials (LJ, EAM, Tersoff, Buckingham, Coulomb) and ensembles (NVT, NPT, NVE, NPH) with optional minimization and shear; create multi-phase workflows (minimize → equilibrate → production) with configurable parameters. * LAMMPS Output Parsing: Parse thermo logs and dump/trajectory files to extract time-series, atom positions, velocities, box dimensions; auto-summarize output files. * Nematic Alignment Analysis: Compute scalar nematic order parameter S, spatial profile S(z), and time evolution S(t) from quaternion data. * Non-Affine Displacement / Plasticity Analysis: Calculate D²min (Falk-Langer) metric to quantify non-affine motion; identify plastic events via thresholding. * Shear Rheology: Generate shear-rate sweep scripts; estimate viscosity via power-law fitting, classifying shear-thinning, Newtonian, or shear-thickening behavior. * DFT / CIF Analysis: Parse, generate, and summarize CIF files, extracting space group, lattice parameters, atomic positions, and symmetry. * MXene Materials Discovery: List, query, compare, and search MXene materials by formula, element, termination, metallic character, band gap, formation energy, etc. * ML Property Prediction: Predict band gap, density, melting point from composition; compute elemental/compositional features; retrieve element properties. * Literature Search: Search arXiv by keyword, author, or materials-science category.

Which integrations are available for this server?

Enables searching arXiv for scientific papers by keyword, with materials-science-optimized search and author-based queries.

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SciMCP Generate a LAMMPS input for a Lennard-Jones fluid under shear at rate 0.01." That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

SciMCP

by Nour-elhaq

Overview Schema Source Related Servers Discussions Admin

Python

Local

SciMCP

Scientific Computing MCP Server for Computational Materials Science

Give your AI coding agent superpowers for molecular dynamics, DFT, crystallography, MXene discovery, ML property prediction, and literature search.

Works with Claude Desktop · Claude Code · Cursor · Windsurf · Any MCP client

uvx scimcp    # Run instantly, no install needed

What is SciMCP?

SciMCP is an MCP server that connects AI assistants to computational materials science workflows. It provides 45 tools across 9 categories: LAMMPS molecular dynamics, DFT/CIF analysis, VASP/QE input generation, MXene materials discovery, ML property prediction, Materials Project queries, phonon analysis, trajectory visualization, and arXiv literature search.

No more copy-pasting LAMMPS scripts. Just tell Claude what you need.

Related MCP server: QCViz-MCP

Quick Start

1. Install

# No install needed (recommended)
uvx scimcp

# Or install with pip
pip install scimcp

# Or with uv
uv add scimcp

2. Connect to your AI client

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "scimcp": {
      "command": "uvx",
      "args": ["scimcp"]
    }
  }
}

claude mcp add scimcp -- uvx scimcp

Add to your MCP config:

{
  "mcpServers": {
    "scimcp": {
      "command": "uvx",
      "args": ["scimcp"]
    }
  }
}

3. Start asking

Generate a LAMMPS input for 10^3 LJ atoms under NVT at 300K with shear

Parse this dump file and show me the temperature evolution

Compute the nematic order for these quaternion data

Run a shear-rate sweep from 0.0001 to 0.05

Parse this CIF file and tell me the space group and lattice parameters

Generate a VASP INCAR for a relaxation calculation

What MXenes are metallic with high conductivity?

Predict the band gap of GaAs

Find recent arXiv papers on MXene battery applications

Search Materials Project for stable oxides

Compute phonon DOS from these frequencies

Plot the thermo dashboard for this simulation

45 Tools — 9 Categories

LAMMPS Input Generation (2 tools)

Tool	What it does
`lammps_generate_input`	Generate complete LAMMPS scripts. Supports LJ, EAM, Tersoff, Buckingham, Coulomb potentials. NVT/NPT/NVE/NPH ensembles. Optional shear deformation.
`lammps_generate_workflow`	Multi-phase workflow: minimize → equilibrate → production. Configurable per-phase ensemble, temperature, and shear.

Output Parsing (3 tools)

Tool	What it does
`lammps_parse_thermo`	Parse thermo logs into structured data with statistics (min/max/mean/std per column).
`lammps_parse_dump`	Parse lammpstrj trajectories. Extract atom positions, velocities, box dimensions per frame.
`lammps_file_summary`	Auto-detect file type (dump vs thermo) and return key metrics.

Nematic Alignment Analysis (3 tools)

Tool	What it does
`lammps_nematic_order`	Scalar order parameter S from quaternion data. S=1 (aligned), S=0 (isotropic). Returns full Q-tensor.
`lammps_nematic_vs_z`	Spatial profile S(z) — alignment as a function of position along z-axis.
`lammps_nematic_vs_time`	Time evolution S(t) — alignment over a trajectory.

Non-Affine Displacement / Plasticity (2 tools)

Tool	What it does
`lammps_d2min`	Falk-Langer D²min metric. Measures how much particle motion deviates from affine deformation of neighbors.
`lammps_identify_plastic_events`	Threshold-based detection of particles that underwent plastic rearrangement.

Shear Rheology (2 tools)

Tool	What it does
`lammps_shear_sweep`	Generate LAMMPS scripts for multiple shear rates. Outputs per-rate input files + config JSON.
`lammps_estimate_viscosity`	Fit power-law model to stress vs rate data. Returns consistency K, flow index n, shear-thinning/thickening classification.

DFT / CIF Analysis (3 tools)

Tool	What it does
`dft_parse_cif`	Parse CIF content and extract space group, lattice parameters, atomic positions, and symmetry data.
`dft_generate_cif`	Generate CIF files from crystallographic parameters (space group, lattice, atom types, fractional coordinates).
`dft_cif_summary`	Quick summary of a CIF file: formula, space group, lattice, element composition.

VASP / Quantum ESPRESSO (5 tools)

Tool	What it does
`dft_vasp_incar`	Generate VASP INCAR files with configurable encut, k-spacing, smearing, spin polarization, and relaxation settings.
`dft_vasp_poscar`	Generate VASP POSCAR files from element lists, fractional coordinates, and lattice parameters. Supports selective dynamics.
`dft_vasp_kpoints`	Generate VASP KPOINTS files with Monkhorst-Pack grids and custom shifts.
`dft_qe_pw_input`	Generate Quantum ESPRESSO pw.x input files for SCF, relaxation, and variable-cell calculations.

MXene Database (5 tools)

Tool	What it does
`materials_mxene_list`	List all 10+ MXenes in the built-in database with key properties.
`materials_mxene_query`	Query by formula, M element, X element, termination, or metallic character.
`materials_mxene_properties`	Get detailed properties for a specific MXene (lattice, band gap, formation energy, applications).
`materials_mxene_compare`	Side-by-side comparison of multiple MXenes.
`materials_mxene_search`	Search MXenes within a property range (band gap, formation energy, elastic modulus).

ML Property Prediction (6 tools)

Tool	What it does
`materials_predict_band_gap`	Predict band gap from composition. Classifies as metal/semiconductor/insulator.
`materials_predict_density`	Estimate material density from composition.
`materials_predict_melting_point`	Estimate melting point from composition.
`materials_predict_all`	Run all prediction models at once.
`materials_element_info`	Get elemental properties (electronegativity, radius, mass, density, melting point).
`materials_composition_features`	Compute ML feature vector (weighted avg properties, EN difference, mixing entropy).

Materials Project (3 tools)

Tool	What it does
`materials_project_query`	Query Materials Project for crystal structures and properties. Uses built-in data when no API key.
`materials_project_details`	Get detailed properties for a specific MP material (mp-id).
`materials_project_stable`	Search for thermodynamically stable materials near the hull.

Phonon Analysis (4 tools)

Tool	What it does
`analysis_phonon_dos`	Compute phonon density of states from frequency list with Gaussian broadening.
`analysis_thermodynamic_properties`	Compute ZPE, Helmholtz free energy, entropy, and heat capacity from phonon frequencies.
`analysis_phonon_band_path`	Generate high-symmetry k-path for phonon band structures (cubic, hexagonal, tetragonal, orthorhombic).
`analysis_phonon_estimate`	Estimate phonon frequency range from composition using mass-force constant model.

Visualization (5 tools)

Tool	What it does
`viz_time_series`	Plot time series data from LAMMPS thermo output.
`viz_histogram`	Plot histograms with statistics (mean, std, median).
`viz_scatter`	Scatter plots, optionally colored by a third variable (e.g., D²min).
`viz_phonon_dos`	Plot phonon density of states with zero-frequency line.
`viz_thermo_dashboard`	Generate 4-panel dashboard (Temperature, Energy, Pressure vs Step).

Literature Search (3 tools)

Tool	What it does
`literature_search_arxiv`	Search arXiv for scientific papers by keyword.
`literature_search_materials`	Materials-science-optimized arXiv search with category filtering.
`literature_search_by_author`	Find papers by a specific author on arXiv.

Example: Full Workflow

from scimcp.tools.lammps.generator import generate_workflow
from scimcp.tools.lammps.shear import write_sweep_scripts, SweepConfig

# Generate a 3-phase simulation with shear
script = generate_workflow(
    atoms_per_side=12,
    potential="lj",
    temperature=300,
    n_production_steps=100000,
    apply_shear=True,
    shear_rate=0.001,
)

# Run a shear-rate sweep
config = SweepConfig(
    shear_rates=[0.0001, 0.001, 0.01, 0.05],
    atoms_per_side=10,
    temperature=300,
)
files = write_sweep_scripts(config)
# Creates: in.shear_0p000100, in.shear_0p001000, in.shear_0p010000, in.shear_0p050000

Sample Output

STEP 5: Nematic Order Parameter S
  Aligned particles:  S = 1.0000
  Random particles:   S = 0.3880
  Partial alignment:  S = 0.4049

  Nematic Profile S(z):
      z_center       S    count
          0.50  1.0000       10   ← aligned
          5.50  0.4483       10   ← transition
          9.50  0.5590       10   ← random

STEP 8: Viscosity Estimation
  Power-law n = 0.80  →  Shear thinning confirmed
  rate=0.0001  →  η=3.15
  rate=0.0100  →  η=1.26
  rate=0.0500  →  η=0.91

Supported Potentials

Potential	LAMMPS Style	Typical Use
Lennard-Jones	`lj/cut`	Noble gases, generic liquids
EAM	`eam/alloy`	Metals (Cu, Al, Ni, Fe, ...)
Tersoff	`tersoff`	Semiconductors (Si, C, Ge)
Buckingham	`buck/coul/long`	Ionic materials
Coulomb	`coul/long`	Charged systems

Architecture

scimcp/
├── src/scimcp/
│   ├── server.py                  # MCP server entry point (45 tools, 7 prompts)
│   └── tools/
│       ├── lammps/
│       │   ├── generator.py       # LAMMPS input generation + workflows
│       │   ├── parser.py          # Dump & thermo file parsing
│       │   ├── nematic.py         # Q-tensor, S(t), S(z) computation
│       │   ├── nonaffine.py       # D²min Falk-Langer metric
│       │   ├── shear.py           # Shear-rate sweep + viscosity fitting
│       │   └── templates/         # Reference LAMMPS templates
│       ├── dft/
│       │   ├── cif.py             # CIF parsing, generation, ASE conversion
│       │   └── vasp_qe.py         # VASP INCAR/POSCAR/KPOINTS + QE pw.x input
│       ├── materials/
│       │   ├── mxene.py           # MXene database (10+ entries)
│       │   ├── prediction.py      # ML property prediction (band gap, density, Tm)
│       │   ├── literature.py      # arXiv literature search
│       │   └── materials_project.py  # Materials Project API + built-in data
│       └── analysis/
│           ├── phonon.py          # Phonon DOS, thermodynamics, band paths
│           └── visualization.py   # Time series, histograms, scatter, dashboards
├── tests/                         # 211 tests (unit + integration)
├── examples/                      # Demo scripts
├── pyproject.toml
└── README.md

Testing

# Run all tests
pytest

# Run with verbose output
pytest -v

# Run specific test module
pytest tests/test_nematic.py -v

# Run Phase 3 tests only
pytest tests/test_vasp_qe.py tests/test_phonon.py tests/test_visualization.py tests/test_materials_project.py -v

# Run integration tests (excludes network-dependent literature tests)
pytest tests/test_phase2_integration.py -v -k "not literature"

211 tests covering all tools: LAMMPS generation/parsing/analysis, CIF handling, VASP/QE input, MXene queries, ML prediction, Materials Project, phonon analysis, visualization, and MCP server integration.

Roadmap

DFT/CIF file parsing and analysis
MXene database queries
ML-accelerated property prediction
ASE (Atomic Simulation Environment) integration
arXiv literature search
Materials Project API integration
Trajectory visualization tools
VASP/QE input generation
Phonon band structure tools

Contributing

Contributions welcome! Whether it's bug fixes, new tools, or documentation.

git clone https://github.com/Nour-elhaq/scimcp.git
cd scimcp
pip install -e ".[dev]"
pytest

License

MIT License — see LICENSE for details.

Built for the computational materials science community

Report Bug · Request Feature

Available Tools

12 tools

lammps_d2minA

Compute non-affine displacement D²min for each particle between two frames.

D²min measures how much a particle's displacement deviates from the best-fit affine deformation of its neighbors. High D²min = plastic event.

Args: positions_t0_json: JSON string, shape (N, 3) positions at time t0. positions_t1_json: JSON string, shape (N, 3) positions at time t1. r_cut: Cutoff radius for neighbor finding.

Inputs: positions_t0_json (required), positions_t1_json (required), r_cut. Returns: result

lammps_estimate_viscosityA

Estimate viscosity from shear rate vs shear stress data.

Fits a power-law model: stress = K * rate^n, where n=1 is Newtonian, n<1 is shear-thinning, n>1 is shear-thickening.

Args: shear_rates_json: JSON array of shear rates. shear_stresses_json: JSON array of corresponding shear stresses.

Inputs: shear_rates_json (required), shear_stresses_json (required). Returns: result

lammps_file_summaryA

Get a summary of a LAMMPS output file (dump or thermo).

Identifies the file type and returns key statistics: number of frames, atoms, columns, box dimensions, or thermo time-series stats.

Args: filepath: Path to the LAMMPS output file.

Inputs: filepath (required). Returns: result

lammps_generate_inputA

Generate a complete LAMMPS input script for molecular dynamics simulation.

Creates a ready-to-run LAMMPS input file with support for LJ, EAM, Tersoff, Buckingham, and Coulomb potentials. Ensembles: NVT, NPT, NVE, NPH. Optional energy minimization and shear deformation.

Args: atoms_per_side: Number of unit cells per side of the cubic box. lattice_spacing: Lattice constant in Angstroms. potential: Interatomic potential type (lj, eam, tersoff, buckingham, coulomb). epsilon: LJ energy parameter. sigma: LJ length parameter. cutoff: Pair potential cutoff distance. ensemble: Thermodynamic ensemble (nvt, npt, nve, nph). temperature: Target temperature. pressure: Target pressure (for NPT/NPH). timestep: Integration timestep. n_steps: Total number of MD steps. dump_freq: Frequency of trajectory dumps. minimize: Run energy minimization before dynamics. apply_shear: Apply shear deformation during dynamics. shear_rate: Shear strain rate. shear_direction: Shear plane (xy, xz, yz). boundary: Boundary conditions. output_file: If provided, write script to this file.

Inputs: sigma, cutoff, epsilon, n_steps, +14 more. Returns: result

lammps_generate_workflowA

Generate a multi-step LAMMPS workflow (minimize → equilibrate → production).

Creates a complete input script with three sequential phases:

Energy minimization (optional)
Equilibration (NVT or NPT)
Production run (with optional shear)

Args: atoms_per_side: Number of unit cells per side. lattice_spacing: Lattice constant. potential: Interatomic potential type. epsilon: LJ energy parameter. sigma: LJ length parameter. cutoff: Pair potential cutoff. timestep: Integration timestep. n_minimize_steps: Number of minimization steps (0 to skip). n_equilibrate_steps: Number of equilibration steps. equilibrate_temp: Temperature for equilibration. equilibrate_ensemble: Ensemble for equilibration. n_production_steps: Number of production steps. production_temp: Temperature for production. production_ensemble: Ensemble for production. production_pressure: Pressure for production (if NPT). apply_shear: Apply shear during production. shear_rate: Shear strain rate. shear_direction: Shear plane (xy, xz, yz). dump_freq: Dump frequency during production. thermo_freq: Thermodynamic output frequency. output_file: Optional file path to write the script.

Inputs: sigma, cutoff, epsilon, timestep, +17 more. Returns: result

lammps_identify_plastic_eventsA

Identify particles that underwent plastic rearrangement.

Args: d2min_json: JSON string with D²min values for each particle. threshold: D²min threshold for plastic events.

Inputs: d2min_json (required), threshold. Returns: result

lammps_nematic_orderA

Compute the scalar nematic order parameter S from quaternion data.

S measures alignment of anisotropic particles. S=1 is perfect alignment, S=0 is isotropic, S=-0.5 is perpendicular alignment.

Input is a JSON array of quaternions [w, x, y, z] for each particle.

Args: quaternions_json: JSON string with shape (N, 4) quaternion data. Example: "[[1,0,0,0], [0.9,0.1,0,0], [0.8,0.2,0.1,0]]"

Inputs: quaternions_json (required). Returns: result

lammps_nematic_vs_timeB

Compute nematic alignment S(t) over a trajectory.

Each frame contains quaternions for all particles at that timestep.

Args: quaternions_frames_json: JSON string — list of (N, 4) arrays, one per frame. Example: "[[[1,0,0,0],[0.9,0.1,0,0]], [[0.8,0.2,0.1,0],[0.7,0.3,0,0]]]"

Inputs: quaternions_frames_json (required). Returns: result

lammps_nematic_vs_zA

Compute nematic alignment profile S(z) along the z-axis.

Bins particles by z-coordinate and computes S in each bin, useful for studying alignment near surfaces or interfaces.

Args: quaternions_json: JSON string with shape (N, 4) quaternion data. z_positions_json: JSON string with shape (N,) z-coordinates. n_bins: Number of bins along z.

Inputs: quaternions_json (required), z_positions_json (required), n_bins. Returns: result

lammps_parse_dumpA

Parse a LAMMPS dump file and return trajectory frame summaries.

Reads lammpstrj files and extracts atom positions, box dimensions, and metadata for each frame.

Args: filepath: Path to the LAMMPS dump file. max_frames: Maximum number of frames to parse (0 = all).

Inputs: filepath (required), max_frames. Returns: result

lammps_parse_thermoA

Parse LAMMPS thermodynamic output and return time-series data.

Reads a LAMMPS log file or thermo.dat file and extracts columns like Step, Temp, PotEng, TotEng, Press, etc.

Args: filepath: Path to the LAMMPS thermo output file.

Inputs: filepath (required). Returns: result

lammps_shear_sweepA

Generate LAMMPS input scripts for a shear-rate sweep.

Creates one input script per shear rate to study rheological behavior, shear thinning/thickening, and viscosity as a function of shear rate.

Args: shear_rates: JSON array of shear rates to simulate. atoms_per_side: Number of unit cells per side. lattice_spacing: Lattice constant. potential: Interatomic potential type. temperature: Simulation temperature. n_equilibrate: Number of equilibration steps. n_production: Number of production steps. shear_direction: Shear plane (xy, xz, yz). output_dir: Directory to write scripts.

Inputs: potential, output_dir, shear_rates, temperature, +5 more. Returns: result

Tool Schema Changelog

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

Loading tool schema changelog…

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.8/5.0

Scored 2026-08-01 03:25 across 12 tools

Disambiguation4/5

Tools are largely distinct in function, but generate_input, generate_workflow, and shear_sweep all produce LAMMPS input scripts, and the three nematic tools share a prefix; descriptions clarify the differences, but a few pairs could be confused at first glance.

Naming Consistency3/5

All tools share the lammps_ prefix, but the suffix mixes verb_noun (generate_input, parse_thermo, identify_plastic_events) with noun phrases (file_summary, nematic_order, shear_sweep). The pattern is readable but not fully consistent.

Tool Count5/5

With 12 tools covering simulation setup, parsing, and analysis, the server is well-scoped and each tool serves a distinct phase of the LAMMPS workflow.

Completeness4/5

The workflow from input generation to trajectory parsing to specialized analysis (nematic order, plastic events, viscosity) is well covered. Minor gaps exist, such as no tool for modifying existing input files or computing other common structural properties, but the core lifecycle is complete.

Maintenance

ActivityInactive

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

MDEngineA
com.forcefieldsilicon
MD workbench for agents: 13 analysis tools, renders, hosted GPU runs; LAMMPS + OpenMM.
Research & Data Data Visualization
141
https://api.forcefieldsilicon.com/mcp
anotherai
dev.anotherai
MCP server for building and testing AI agents with multi-model experimentation and insights.
Agent Orchestration Autonomous Agents Testing & QA Tools
2
https://api.anotherai.dev/mcp
Wauldo
com.wauldo
Stateless agentic tools over MCP: concept extraction, long-context, knowledge graph, planning.
Knowledge & Memory Autonomous Agents
2
https://api.wauldo.com/mcp
Arquestra — Scientific Computing Orchestration
io.github.Apertura-AI
AI orchestration for computational chemistry and HPC workflows.
Bioinformatics AI & Machine Learning
2
https://api.apertura-ai.de/mcp/orchestration

Related MCP Servers

MCP LAMMPS Server
Bioinformatics Autonomous Agents
Chenghao-Wu
A
license
Not graded
quality
F
maintenance
A Model Context Protocol server that enables AI assistants to interact with LAMMPS for molecular dynamics simulations through natural language commands.
Updated a year ago (2025-12-27 02:44 UTC)
13
Apache 2.0
QCViz-MCP
dydtkddl
A
license
Not graded
quality
C
maintenance
An MCP server for quantum chemistry that enables LLMs to perform electronic structure analysis, parse calculation outputs, and generate 3D orbital visualizations. It integrates tools like PySCF, cclib, and py3Dmol to facilitate molecular structure manipulation and bonding analysis through natural language.
Updated 7 months ago (2026-04-03 05:38 UTC)
BSD 3-Clause
colabfit-mcpofficial
Research & Data AI & Machine Learning
colabfit
F
license
Not graded
quality
D
maintenance
An MCP server for discovering ColabFit materials science datasets and training MACE interatomic potentials on local hardware, enabling AI assistants to search, download, train, and validate models.
Updated 3 months ago (2026-08-14 12:23 UTC)
3
-
fairchem-mcp
AI & Machine Learning Research & Data Autonomous Agents
jkitchin
A
license
A
quality
C
maintenance
An interactive MCP server for FAIRChem and ASE simulations that allows LLM agents to load a model once and steer relaxations, MD, NEB, phonons, and minima searches mid-flight, with live monitoring and code introspection.
Updated 17 days ago (2026-10-27 18:09 UTC)
17
1
MIT

SciMCP

Scientific Computing MCP Server for Computational Materials Science

What is SciMCP?

Quick Start

1. Install

2. Connect to your AI client

3. Start asking

45 Tools — 9 Categories

LAMMPS Input Generation (2 tools)

Output Parsing (3 tools)

Nematic Alignment Analysis (3 tools)

Non-Affine Displacement / Plasticity (2 tools)

Shear Rheology (2 tools)

DFT / CIF Analysis (3 tools)

VASP / Quantum ESPRESSO (5 tools)

MXene Database (5 tools)

ML Property Prediction (6 tools)

Materials Project (3 tools)

Phonon Analysis (4 tools)

Visualization (5 tools)

Literature Search (3 tools)

Example: Full Workflow

Sample Output

Supported Potentials

Architecture

Testing

Roadmap

Contributing

License

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

MCP LAMMPS Server

QCViz-MCP

colabfit-mcpofficial

fairchem-mcp