Skip to content

Latest commit

 

History

History
61 lines (46 loc) · 2.21 KB

File metadata and controls

61 lines (46 loc) · 2.21 KB

name: 'chemcrow-drug-discovery' description: 'An LLM chemistry agent with expert-designed tools for organic synthesis, drug discovery, and materials design.' measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools:

  • read_file
  • run_shell_command

ChemCrow

ChemCrow is an open-source package for the accurate integration of Large Language Models (LLMs) with chemistry tools. It is designed to autonomously plan and execute chemical syntheses, research materials, and discover new drugs.

When to Use This Skill

  • Synthesis Planning: "How do I synthesize Ibuprofen?"
  • Property Prediction: "What is the logP of this molecule?"
  • Safety Checks: "Is this reaction explosive?"
  • Literature Search: "Find patents related to this substructure."

Core Capabilities

  1. Synthesis Planning: Uses tools like RXN4Chemistry or local retrosynthesis models to plan routes.
  2. Molecule Manipulation: Add/remove functional groups, generate SMILES (RDKit).
  3. Safety Assessment: Checks reagents against safety databases (PubChem, GHS).
  4. Web Search: Google/Patent search for chemical literature.

Tools (Expanded List - v2.0)

  • RDKit: Cheminformatics (MolToSmiles, SmilesToMol, descriptors).
  • PubChem: Search for properties and safety data.
  • LiteratureSearch: Search arXiv, patents.
  • Synthesis: IBM RXN, OPLS.
  • Name2SMILES: Convert chemical names to structures.

Example Usage

User: "Plan a synthesis for a derivative of Aspirin that might have better solubility."

Agent Action:

  1. Retrieves Aspirin structure (SMILES).
  2. Modifies structure (e.g., adds a polar group) using RDKit.
  3. Checks solubility prediction.
  4. Plans synthesis route for the new molecule.
  5. Checks safety of reagents.