Molecular Dynamics (Claude Skill)

Claude skill that sets up, runs, and analyzes molecular dynamics simulations end-to-end — OpenMM for the simulation engine and MDAnalysis for trajectory post-processing.

   
Type Claude Skill
Supplier K-Dense Inc.
Availability GA — distributed via the K-Dense scientific-agent-skills collection
Pricing Free / OSS (skill source); OpenMM is MIT, MDAnalysis is GPLv2+
Capabilities Read/Write — Claude writes and runs Python locally (system preparation, MD execution, trajectory analysis)

How to install

  • Claude Code / Claude.ai — Skills CLI (recommended): 
    npx skills add K-Dense-AI/scientific-agent-skills

    Installs the K-Dense collection; enable the molecular-dynamics skill when prompted (also works in Cursor/Codex via the Agent Skills spec; requires Node ≥ 18).

  • Claude Code / Claude Desktop — manual clone: 
    git clone https://github.com/K-Dense-AI/scientific-agent-skills
cp -r scientific-agent-skills/skills/molecular-dynamics ~/.claude/skills/
pip install openmm MDAnalysis

    Project-scoped alternative: copy into .claude/skills/ (inside your project) instead of ~/.claude/skills/.

What it does

SKILL.md with recipes for:

  • System preparation — protein and small-molecule loading, solvation, ion placement, force-field assignment (AMBER, CHARMM, OPLS via OpenMM-compatible parameter sets)
  • Simulation — energy minimization, NVT/NPT equilibration, production MD with checkpointing
  • Trajectory analysis (MDAnalysis) — RMSD, RMSF, radius of gyration, contact maps, hydrogen-bond analysis, free-energy surfaces via PCA / order parameters
  • Visualization hooks — outputs PDB / DCD / XTC that downstream PyMOL or ChimeraX sessions can consume (see also Molecule-MCP)

Primary use cases: protein-stability assessment, ligand-binding-pose refinement and MM/PBSA-style scoring, conformational-ensemble generation for drug-discovery targets, biophysics teaching workflows.

Notes

  • Compute-heavy. Production MD assumes a CUDA-capable GPU; OpenMM’s CPU platform works for small systems and tutorials. Cloud GPU runs are the typical pattern.
  • Skill is documentation plus Python recipes — Claude executes OpenMM and MDAnalysis locally via Bash/Python. The skill does not ship binaries.
  • Complements the AlphaFold MCP Server and PDB MCP Server for structure retrieval, and Molecule-MCP for PyMOL/ChimeraX/GROMACS visualization and execution.

Sources

Installed this tool?

Share feedback — install path, OS, errors, workarounds. The form opens with this tool pre-selected and a link back to this page.