Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/learningmatter-mit/atomisticskills/drug-complex-system-buildernpx skills add learningmatter-mit/AtomisticSkills --skill drug-complex-system-buildergit clone --depth 1 https://github.com/learningmatter-mit/AtomisticSkillsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00118 | $0.01978 |
| Opus 5 | $0.00059 | $0.00989 |
| Sonnet 5 | $0.00024 | $0.00396 |
| Haiku 4.5 | $0.00012 | $0.00198 |
Grade A, and why
drug-complex-system-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
drug-complex-system-builder
Goal
To take a prepared protein (PDB) and a validated ligand pose (SDF) and produce a fully parameterized, solvated, ion-neutralized OpenMM simulation bundle ready for drug-protein-ligand-md.
The output bundle includes:
- Serialized OpenMM System XML (force field parameters, constraints)
- Full-precision initial state XML (positions + box vectors for exact restart)
- Solvated PDB with protein + ligand + water + ions (for visualization)
- Provenance JSON recording all build parameters
Instructions
1. Prepare inputs
Required inputs:
- Receptor PDB: from drug-protein-prep (protonated, missing residues resolved).
- Ligand SDF: from drug-pose-validation or drug-docking-vina. Must have 3D coordinates in the receptor frame and explicit hydrogens.
2. Build the solvated complex
# Env: drugmd-agent
python .agents/skills/drug-complex-system-builder/scripts/build_complex.py \
--receptor docking/inputs/protein_prepared.pdb \
--ligand docking/validation/valid_poses.sdf \
--ligand_ff openff-2.2.0 \
--protein_ff amber/ff14SB \
--water_model tip3p \
--box_padding 12.0 \
--ionic_strength 0.15 \
--output_dir md/system/
Key parameters:
--ligand_ff: Force field for the ligand. Options:openff-2.2.0(Sage, recommended),gaff-2.11. OpenFF Sage is generally preferred for drug-like molecules.--protein_ff: Protein force field. Default:amber/ff14SB.--water_model: Water model. Default:tip3p. Options:tip3p,tip3pfb,tip4pew,opc,spce. Usetip3pfboropcfor better accuracy at higher cost.--box_padding: Minimum distance from solute to box edge in Angstroms (default: 12.0). Use 10-12 A for production; smaller values risk periodic image artifacts.--ionic_strength: Target NaCl concentration in mol/L (default: 0.15, physiological). The system is always charge-neutralized first; additional ion pairs are added to reach the target ionic strength. The ionic strength calculation does not count the neutralization ions (they are treated as bound to the solute).--pose_index: Which pose from the SDF to use (default: 0, the top-ranked pose).--box_shape: Simulation box geometry (default:cube). Options:cube,dodecahedron,octahedron. Dodecahedron and octahedron use ~30% less water for the same minimum solute-edge distance.--hydrogen_mass: Hydrogen mass in amu for hydrogen mass repartitioning (default: 4.0). With HMR (3-4 amu), the script usesAllBondsconstraints, enabling 4-5 fs timesteps (OpenMM recommends 5 fs withLangevinMiddleIntegrator). Set to 1.008 to disable HMR (usesHBondsconstraints, requires 2 fs timestep). Note: at 4 amu, methyl carbons become lighter than their bonded hydrogens, which can affect dynamics in some systems (particularly membranes). Use 3 amu if this is a concern. The downstream MD skill must use a matching timestep (checkhmr_enabledandconstraintsin the provenance JSON).
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 112 lines · 118 tokens per session scan A 7f27a693ddb5
drug-complex-system-builder is a skill published in the GitHub repository learningmatter-mit/AtomisticSkills (154 stars, last pushed 7d ago), licensed MIT. It adds 118 tokens to every session and 1,978 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
datamol
Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters…
deepchem
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first…
binding-affinity
Hybrid ML + physics binding affinity prediction. Empirical scoring, MM/GBSA rescoring, multi-method consensus, and batch virtual screening for protein-ligand complexes.
drug-design
End-to-end drug discovery pipeline orchestration. Deterministic Python script that auto-chains structure prediction, pocket detection, de novo design, docking, scoring, and ADMET filtering into reproducible workflows.
molecular-optimization
Iterative lead optimization with analyze-reason-generate-verify-evaluate loop. Paper-backed (MT-Mol, DrugR, MultiMol).
admet-reasoning
Interpretable ADMET analysis with mechanistic reasoning. Maps liabilities to structural causes and biological pathways. Based on CoTox (Park 2025) and DrugR (Liu 2026).