# Diffdock

> Diffdock is an Agent Skill for general use published in the companion-inc/feynman repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Run or plan DiffDock molecular docking workflows. Use when a task asks for protein-ligand pose prediction, docking setup, ligand/protein preparation, pose ranking, or docking-result verification.

- Skill name: `diffdock`
- Category: [Others](https://www.openparable.com/skill/others/1)
- Source: https://github.com/companion-inc/feynman/blob/HEAD/skills/diffdock/SKILL.md
- Repository: companion-inc/feynman (8,301 GitHub stars when collected)

## SKILL.md

# DiffDock

Use this skill for protein-ligand docking and pose review.

Workflow:

1. Record protein source, chain selection, binding site context, ligand identity, protonation/tautomer assumptions, and known cofactors.
2. Verify the available execution path and dependency stack before claiming a docking run is possible.
3. Preserve input PDB/mmCIF, ligand SDF/SMILES, prepared structures, command, seed, package version, and logs.
4. Save ranked poses, confidence scores, contact summaries, and 3D previews as Feynman artifacts.
5. Compare poses against known ligands, active-site residues, experimental structures, or orthogonal docking where the conclusion matters.

Report docking as a ranked hypothesis, not binding proof.

---

Source: https://www.openparable.com/skill/others/diffdock (OpenParable). Cite the canonical URL above.