ClawCures
An AI Agent for Drug Discovery
ClawCures is an experimental AI agent built on OpenClaw. It plans and runs drug discovery research, starting with severe and neglected diseases and looking for treatment approaches supported by the biology.
Find cures for all diseases by prioritizing the highest-burden conditions and researching the best drug design strategies for each.
The default ClawCures prompt
What Is ClawCures?
ClawCures can work on a specific disease or research question. Its default objective is broader: prioritize severe and neglected diseases, then investigate possible treatments.
The agent reviews scientific literature, runs machine-learning models, designs and ranks compounds, coordinates experiments with contract research organizations (CROs), and drafts regulatory documents such as IND applications.
The code is open source, so you can inspect how it works and try it with your own research objectives.
Key Capabilities
- Reads and synthesizes scientific literature
- Runs ML models for drug design
- Designs de novo antibodies with optimized properties
- Coordinates CRO wet-lab experiments
- Prepares IND regulatory documentation
- Prioritizes highest-burden diseases
See It In Action
Start ClawCures with one command. Give it a disease to investigate, or use its default objective.
$ clawcures run
🦞 Initializing ClawCures autonomous agent...
📋 Loading default objective: cure all human disease
🔬 Ranking disease portfolio by global burden...
🧬 Generating therapeutic hypotheses...
⚡ Planning via OpenClaw gateway...
✅ Campaign running. Iteration 1 of ∞
How It Works
ClawCures separates planning from execution. OpenClaw generates strategic plans, while Refua MCP executes typed scientific tools in an auditable workflow.
Define Objective
Set a disease target or let ClawCures default to its all-disease mission.
Literature Synthesis
The agent reviews scientific papers and compares possible treatment strategies.
Compound Design
ML models design and evaluate small molecules, peptides, and new antibodies.
Lab Coordination
ClawCures coordinates CRO wet-lab experiments to validate promising candidates.
Regulatory Prep
Draft IND sections and supporting documents for review.
From Research to Documentation
ClawCures connects computational modeling, lab coordination, and regulatory document preparation.
Literature Review
Reads and indexes scientific papers to identify treatment ideas and the evidence behind them.
ML-Powered Design
Runs machine-learning models to design and score compounds, predict binding affinity, and optimize drug-likeness.
De Novo Antibodies
Designs antibody candidates computationally, considering predicted binding, stability, and suitability for development.
Wet-Lab Coordination
Connects with CRO partners to commission lab experiments and track their results.
Regulatory Documentation
Drafts IND applications, study reports, and regulatory dossiers from the research record.
Disease Prioritization
Ranks diseases by severity and neglect to help choose where to focus research.
Part of the AgentCures Ecosystem
ClawCures is part of Refua, the AgentCures collection of open-source tools for drug discovery.
It brings together Amiti for molecular screening, Refua MCP for scientific tools, and OpenClaw for planning, alongside tools for clinical simulation, regulatory evidence, and data.
You can add integrations to connect the agent with the services your research team uses.
Ecosystem Integration
- Amiti molecular embedding model
- Refua MCP scientific tool server
- OpenClaw planning engine
- Clinical trial design tools
- Regulatory dossier generation
- Extensible integration framework
Join the Mission
Try ClawCures with a research question, inspect its plans, and help improve the tools. The project is open source and welcomes contributions.