Molecular Viz Agent Skill
Visualize drug-protein complexes using build_viewer.py, PubChem, and OpenFold3 NIM. Use when asked to show a molecular structure, drug target, or protein visualization.
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/NVIDIA/dgx-spark-playbooks --skill molecular-vizThe instruction itself
8 sections, as written by the authorMolecular Visualization
Generate 3D protein-ligand visualizations using the build_viewer.py script. The script handles the full pipeline:
- Drug SMILES -- looked up automatically from PubChem
- Protein target -- resolved from a built-in drug-target table (or pass
--sequencemanually) - Structure prediction -- protein + drug sent to OpenFold3 NIM for co-structure prediction
- 3D viewer -- self-contained HTML with jQuery + 3Dmol.js inlined, saved to canvas
Usage
Simplest form (target auto-resolved):
python /sandbox/clinical-intelligence/scripts/build_viewer.py --drug metformin
With explicit sequence (for drugs not in the built-in table):
python /sandbox/clinical-intelligence/scripts/build_viewer.py --drug drugname --sequence AMINOACIDSEQ --title "Custom Title"
Options
| Flag | Required | Description |
|------|----------|-------------|
| --drug | Yes | Drug name for PubChem SMILES lookup (e.g. metformin) |
| --sequence | No | Amino acid sequence of protein target. Auto-resolved if omitted. |
| --title | No | Custom viewer title |
| --output | No | Custom output path (defaults to ~/.openclaw/canvas/{drug}_complex.html) |
| --openfold-host | No | Override OpenFold3 host IP (defaults to 172.17.0.1) |
Built-in drug targets
The script knows these drugs and auto-resolves their protein targets:
| Drug | Target protein |
|------|---------------|
| metformin | Insulin B-chain |
| atorvastatin | HMG-CoA reductase |
| rosuvastatin | HMG-CoA reductase |
| lisinopril | ACE |
| enalapril | ACE |
| losartan | Angiotensin II receptor type 1 |
| amlodipine | L-type calcium channel Cav1.2 |
| empagliflozin | SGLT2 |
| semaglutide | GLP-1 receptor |
For any drug in this table, just pass --drug and the script does the rest.
Drugs NOT in the table
If the drug is not listed, the script exits with an error and prints the list of known drugs. In that case, you need to provide --sequence explicitly. Tell the user the drug is not in the built-in table and that you need a protein target sequence to proceed.
Drugs that cannot be visualized
Biologics, enzyme mixtures, or complex formulations that PubChem cannot resolve to a single SMILES (e.g. pancrelipase, insulin glargine) will still get protein-only structure prediction -- the script handles this gracefully by predicting without a ligand.
Output
The script saves an HTML viewer to canvas. Link it in your response as a markdown hyperlink:
[View 3D structure](http://localhost:18789/__openclaw__/canvas/metformin_complex.html)
Confidence Scores
The viewer header displays OpenFold3 scores:
- Confidence -- overall prediction confidence (higher = better)
- pLDDT -- per-residue local confidence (0-100, >70 is good)
- pTM -- predicted template modeling score (0-1)
- ipTM -- interface predicted TM-score (complexes only, measures protein-ligand interface quality)
Other skills for the same job
different authors, same section of the catalogue
Comprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization. When Claude needs to work with spreadsheets (.xlsx, .xlsm, .csv, .tsv, etc) for: (1) Creating new spreadsheets with formulas and formatting, (2) Reading or analyzing data, (3) Modify existing spreadsheets while preserving formulas, (4) Data analysis and visualization in spreadsheets, or (5) Recalculating formulas
Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.
Picks random winners from lists, spreadsheets, or Google Sheets for giveaways, raffles, and contests. Ensures fair, unbiased selection with transparency.
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
MATLAB and GNU Octave numerical computing for matrix operations, data analysis, visualization, and scientific computing. Use when writing MATLAB/Octave scripts for linear algebra, signal processing, image processing, differential equations, optimization, statistics, or creating scientific visualizations. Also use when the user needs help with MATLAB syntax, functions, or wants to convert between MATLAB and Python code. Scripts can be executed with MATLAB or the open-source GNU Octave interpreter.
UMAP dimensionality reduction. Fast nonlinear manifold learning for 2D/3D visualization, clustering preprocessing (HDBSCAN), supervised/parametric UMAP, for high-dimensional data.
Creating interactive data visualisations using d3.js. This skill should be used when creating custom charts, graphs, network diagrams, geographic visualisations, or any complex SVG-based data visualisation that requires fine-grained control over visual elements, transitions, or interactions. Use this for bespoke visualisations beyond standard charting libraries, whether in React, Vue, Svelte, vanilla JavaScript, or any other environment.
Access AlphaFold 200M+ AI-predicted protein structures. Retrieve structures by UniProt ID, download PDB/mmCIF files, analyze confidence metrics (pLDDT, PAE), for drug discovery and structural biology.
How to use it
Take nvidia/molecular-viz from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.