Chemistry Query Agent for Openclaw

A comprehensive chemistry toolkit for PubChem data retrieval and RDKit molecular analysis designed for AI agents.

cheminem
v1.0.0
Feb 17, 2026
0
2.3k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install chemistry-query

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install chemistry-query using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Chemistry Query Agent?

The Chemistry Query agent skill is a full-stack chemistry toolkit that bridges the gap between raw chemical data and actionable insights. By combining the PubChem REST API for data retrieval with the RDKit cheminformatics engine, this skill allows AI agents to perform complex molecular analysis, visualization, and synthesis planning. It is a vital component for researchers and developers building sophisticated scientific workflows using Openclaw Skills.

This skill enables agents to resolve chemical names to structures, calculate critical molecular properties like logP and TPSA, and even perform recursive retrosynthesis. Whether you are looking for specific compound references or simulating forward reactions with SMARTS templates, this skill provides the structured JSON output necessary for seamless multi-agent collaboration.

Chemistry Query Agent Use Cases

  • Automating compound lookups to retrieve IUPAC names, formulas, and InChIKeys.
  • Generating high-quality 2D PNG or SVG visualizations for molecular structures.
  • Conducting recursive BRICS retrosynthesis to identify potential synthetic precursors.
  • Calculating molecular descriptors and fingerprints for similarity or substructure searches.
  • Planning multi-step synthesis routes using established reaction templates.

How Chemistry Query Agent Works

  1. The user provides a chemical name, CID, or SMILES string to the agent.
  2. The skill utilizes query_pubchem.py to resolve names and fetch metadata from the PubChem API.
  3. The RDKit engine processes the molecular data to calculate properties or generate structural diagrams.
  4. Advanced logic performs retrosynthetic disconnects or simulates reactions based on SMARTS templates.
  5. The results are packaged into a unified JSON schema, making it compatible with other Openclaw Skills for downstream processing.

Chemistry Query Agent Setup

To integrate this skill into your environment, ensure you have RDKit 2024.03+ installed. Use the following commands to initialize queries via the CLI:

# Query PubChem for compound information
python scripts/query_pubchem.py --compound "aspirin" --type info

# Generate molecular properties from a SMILES string
python scripts/rdkit_mol.py --smiles "CC(=O)Oc1ccccc1C(=O)O" --action props

# Execute a full chain analysis (name to properties and visualization)
python scripts/chain_entry.py --input-json '{"name": "caffeine", "context": "user"}'

Chemistry Query Agent Data Schema & Taxonomy

The skill returns a structured JSON object containing detailed chemical metadata and analysis. This format is optimized for integration with other Openclaw Skills.

Key Type Description
smiles string The canonical SMILES representation of the molecule.
report object Contains sub-objects for props (MW, logP), draw (file paths), and retro.
viz array A list of file paths to generated PNG or SVG visualizations.
confidence float A score representing the reliability of the name resolution or calculation.
timestamp string ISO8601 formatted string of the execution time.

Chemistry Query Agent Advanced Features

  • Support for 21 named reaction templates including Suzuki, Heck, and Diels-Alder for realistic reaction simulation.
  • Multi-step synthesis planning with recursive depth control for complex molecule disassembly.
  • 3D coordinate generation and MMFF optimization for molecular modeling.
  • Morgan fingerprinting and Tanimoto similarity scoring for chemical library screening.
  • Automated name-to-structure resolution across all RDKit-driven actions.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*