TorchDrug Workflow Assistant
Writes and debugs TorchDrug 0.2.1 code for molecular property prediction, generation, retrosynthesis, protein encoders, and knowledge graph reasoning.
Dev & CodingAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026
What it does
- Runs a version guard first (Python 3.7–3.10, PyTorch 1.8–2.0; Apple Silicon is CPU-only, no MPS) so unverified combos aren't presented as supported.
- Generates code following the canonical four-step pattern:
datasets.*→models.*→tasks.*→core.Engine. - Covers official workflows for property prediction, self-supervised pretraining (InfoGraph, AttributeMasking), molecule generation (GCPN, GraphAF), retrosynthesis (USPTO50k), KG reasoning (RotatE, NeuralLP), and protein modeling (ESM, GearNet).
- Includes a troubleshooting section for install/import failures, feature-dimension mismatches, device mismatches, and checkpoint loading.
Who it's for
- Drug-discovery, cheminformatics, and protein-modeling researchers working in PyTorch.
- Anyone burned by LLMs inventing nonexistent TorchDrug APIs or method names.
- Teams who need reproducible torch-scatter / torch-cluster wheel pinning.
Example uses
- "Build a GIN property-prediction script on ClinTox" → 80/10/10 split, AUPRC/AUROC metrics, Engine training loop.
- "Set up a retrosynthesis pipeline on USPTO50k" → synchronized reaction and synthon dataset views, separate CenterIdentification and SynthonCompletion training, then combined via
tasks.Retrosynthesis. - "
import torchdrugfails with a torch-scatter error" → checks Python/PyTorch/CUDA/ABI as one compatibility set and suggests a reproducible CPU wheel combination.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/torchdrug/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the skills/torchdrug folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/torchdrug/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/torchdrug ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repo:
git clone https://github.com/K-Dense-AI/scientific-agent-skills.git - Create the skills folder if needed:
mkdir -p ~/.claude/skills - Copy just this skill:
cp -r scientific-agent-skills/skills/torchdrug ~/.claude/skills/ - Verify the bundled
references/markdown files came along. - Restart Claude Code, then ask something like "Write TorchDrug code to train GIN on ClinTox" and the skill will activate.
- To actually run code, create a Python 3.10 env and install
torch==2.0.0, matchingtorch-scatter/torch-clusterwheels, thentorchdrug==0.2.1.
View source on GitHub ↗License: MIT