Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Combines PubChem compound lookup with RDKit cheminformatics to retrieve molecular properties, generate 2D structure images, and plan synthesis routes. Accepts compound names or SMILES strings and outputs structured JSON. Includes 21 named reaction templates and BRICS-based retrosynthesis with configurable depth.
Chains PubChem name resolution with RDKit processing in a single call, eliminating manual API queries and separate cheminformatics scripts.
When to use it
- Looking up molecular weight and logP for a drug candidate by name
- Generating a 2D structure image of a SMILES string for a report
- Planning retrosynthesis routes for a target pharmaceutical compound
- Simulating a Suzuki coupling reaction with custom reactants
- Finding structurally similar compounds to a known molecule by fingerprint
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: What are the molecular properties and retrosynthesis routes for caffeine?
- 1Resolves 'caffeine' to a canonical SMILES string via PubChem
- 2Computes molecular properties: MW, logP, TPSA, HBD, HBA, rotatable bonds
- 3Runs BRICS retrosynthesis to depth 2 and returns precursor fragments
- 4Generates a 2D PNG structure image
- 5Returns unified JSON with props, retro fragments, and image path
Structured JSON with molecular properties, BRICS precursors, and a path to the generated PNG