Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Run and analyze molecular dynamics simulations with OpenMM and MDAnalysis. Set up protein/small molecule systems, define force fields, run energy minimization and production MD, analyze trajectories (RMSD, RMSF, contact maps, free energy surfaces). For structural biology, drug binding, and biophysics.
.claude/skills/mkurman-molecular-dynamics/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 68% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 36% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 30% | 0% |
--|------------------------|-------------| | Standard proteins | AMBER14 (amber14-all.xml) | TIP3P-FB | | Proteins + small molecules | AMBER14 + GAFF2 | TIP3P-FB | | Membrane proteins | CHARMM36m | TIP3P | | Nucleic acids | AMBER99-bsc1 or AMBER14 | TIP3P | | Disordered proteins | ff19SB or CHARMM36m | TIP3P |
pythonfrom pdbfixer import PDBFixer from openmm.app import PDBFile def fix_pdb(input_pdb, output_pdb, ph=7.0): """Fix common PDB issues: missing residues, atoms, add H, standardize.""" fixer = PDBFixer(filename=input_pdb) fixer.findMissingResidues() fixer.findNonstandardResidues() fixer.replaceNonstandardResidues() fixer.removeHeterogens(True) # Remove water/ligands fixer.findMissingAtoms() fixer.addMissingAtoms() fixer.addMissingHydrogens(ph) with open(output_pdb, 'w') as f: PDBFile.writeFile(fixer.topology, fixer.positions, f) return output_pdb
python# For ligand parameterization, use OpenFF toolkit or ACPYPE # pip install openff-toolkit from openff.toolkit import Molecule, ForceField as OFFForceField from openff.interchange import Interchange def parameterize_ligand(smiles, ff_name="openff-2.0.0.offxml"): """Generate GAFF2/OpenFF parameters for a small molecule.""" mol = Molecule.from_smiles(smiles) mol.generate_conformers(n_conformers=1) off_ff = OFFForceField(ff_name) interchange = off_ff.create_interchange(mol.to_topology()) return interchange
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 11,817 | 10,889 | -8% | 1 | 1 | 0% | 2,703 | 3,173 | +17% | 0 | 0 | — |
case-02 | pass→pass | 17,848 | 17,808 | -0% | 1 | 1 | 0% | 2,972 | 4,043 | +36% | 0 | 0 | — |
case-03 | fail→fail | 11,817 | 10,176 | -14% | 1 | 1 | 0% | 2,319 | 3,044 | +31% | 0 | 0 | — |
case-04 | fail→pass | 13,750 | 11,320 | -18% | 1 | 1 | 0% | 2,235 | 2,957 | +32% | 0 | 0 | — |
case-05 | fail→pass | 13,433 | 11,821 | -12% | 1 | 1 | 0% | 2,237 | 3,042 | +36% | 0 | 0 | — |
case-06 | pass→pass | 14,429 | 12,873 | -11% | 1 | 1 | 0% | 2,371 | 3,071 | +30% | 0 | 0 | — |
case-07 | fail→fail | 17,598 | 17,988 | +2% | 1 | 1 | 0% | 2,821 | 3,946 | +40% | 0 | 0 | — |
case-08 | pass→pass | 7,164 | 5,343 | -25% | 1 | 1 | 0% | 1,258 | 1,753 | +39% | 0 | 0 | — |
case-09 | fail→fail | 13,857 | 13,309 | -4% | 1 | 1 | 0% | 2,419 | 3,194 | +32% | 0 | 0 | — |
case-10 | pass→fail | 12,520 | 14,676 | +17% | 1 | 1 | 0% | 1,929 | 3,242 | +68% | 0 | 0 | — |
case-11 | pass→pass | 11,042 | 11,579 | +5% | 1 | 1 | 0% | 1,616 | 2,675 | +66% | 0 | 0 | — |
case-12 | pass→pass | 11,734 | 16,232 | +38% | 1 | 1 | 0% | 1,963 | 3,488 | +78% | 0 | 0 | — |
case-13 | pass→pass | 16,636 | 15,976 | -4% | 1 | 1 | 0% | 2,575 | 3,451 | +34% | 0 | 0 | — |
case-14 | pass→pass | 15,266 | 17,637 | +16% | 1 | 1 | 0% | 2,498 | 4,000 | +60% | 0 | 0 | — |
case-15 | pass→pass | 14,629 | 14,472 | -1% | 1 | 1 | 0% | 2,444 | 3,258 | +33% | 0 | 0 | — |
case-16 | pass→pass | 6,945 | 5,930 | -15% | 1 | 1 | 0% | 1,208 | 1,822 | +51% | 0 | 0 | — |
case-17 | pass→pass | 6,706 | 4,417 | -34% | 1 | 1 | 0% | 1,141 | 1,698 | +49% | 0 | 0 | — |
case-18 | pass→pass | 5,948 | 3,835 | -36% | 1 | 1 | 0% | 960 | 1,431 | +49% | 0 | 0 | — |
case-19 | pass→pass | 6,252 | 7,130 | +14% | 1 | 1 | 0% | 1,007 | 1,898 | +88% | 0 | 0 | — |
case-20 | fail→fail | 27,974 | 13,268 | -53% | 1 | 1 | 0% | 1,351 | 3,215 | +138% | 0 | 0 | — |
case-21 | fail→fail | 15,159 | 15,314 | +1% | 1 | 1 | 0% | 2,703 | 3,716 | +37% | 0 | 0 | — |
case-22 | fail→fail | 12,144 | 14,306 | +18% | 1 | 1 | 0% | 2,158 | 3,475 | +61% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +5 percentage points is the difference between those two pass rates over the 21 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.