Open toxicology tools for AI assistants, built so people can inspect the evidence, trace the sources, and see the uncertainty.
Each module uses the Model Context Protocol (MCP), an open standard that lets compatible AI assistants call external tools. Use one module on its own, connect several into a workflow, or explore the code and evidence without an AI assistant.
- Browse the suite guide to understand the tools and choose a starting point.
- Start with CompTox for source-linked chemical identity and screening evidence from the U.S. Environmental Protection Agency.
- Read the preprint for the scientific background.
| If you want to explore… | Start here |
|---|---|
| Chemical identity and screening evidence | CompTox MCP |
| Exposure scenarios | Direct-Use Exposure MCP |
| Environmental fate | Environmental Fate MCP |
| Kinetics and internal dose | Physiologically Based Pharmacokinetic (PBPK) MCP |
| Mechanistic pathways | Adverse Outcome Pathway (AOP) MCP |
| Chemical grouping and read-across | OECD QSAR Toolbox MCP, for quantitative structure–activity relationships |
| Rapid screening of absorption, distribution, metabolism, excretion, and toxicity | ADMETlab MCP |
- Evidence you can trace: results keep source links and provenance close at hand.
- Boundaries you can see: assumptions, limitations, and uncertainty stay visible.
- Pieces you can combine: structured results can move between compatible tools.
- People stay in control: screening evidence remains clearly separated from stronger scientific conclusions.
These tools support research and screening. A runnable model or returned result is not automatically scientifically qualified, clinically appropriate, or ready for a regulatory decision. Review important outputs independently and follow each upstream data provider's access, rate-limit, license, and attribution terms.
Contributions are welcome. Start with the repository that owns the tool you want to improve and read its local guidance. Organization-wide defaults are available in our contributing guide and security policy.
ToxMCP was developed in part through the VHP4Safety project and related computational-toxicology research. Funding included the Dutch Research Council (NWO), grant NWA.1292.19.272.