ClawBio plugin
ClawBio is a library of 110 agent skills for bioinformatics: genomics, pharmacogenomics and population genetics, reproducible.
- 1,182 ★ on the repository
- Last commit
Extended access
Why this level
- The skills run Python analyses over genomic files and call public databases.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/clawbioDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skills run Python analyses over genomic files and call public databases.
This is third-party code. Review the repository files before installing.
What it does
ClawBio collects Agent Skills for bioinformatics tasks. The `skills/` directory holds separate folders: FASTA analysis, clinical variant prioritization, CNV classification by ACMG, clinical trial search, data fetching from ENA and ArrayExpress, pharmacogenomics from a photo of drug packaging and others, 110 in total by `skills/catalog.json`. Install with `pip install clawbio` and run `clawbio run pharmgx --demo`; the package is also on bioconda. For Claude Code there is a plugin: `/plugin marketplace add ClawBio/ClawBio`. Cursor, VS Code, Codex and Zed read the same folders from `~/.agents/skills/` or the project `.agents/skills/`. The project runs locally, and many analyses write command replays, environment details and result checksums to a `reproducibility/` folder. The MCP server is deprecated since version 0.7.0 and removed in 0.8.0.
Who it is for. Bioinformaticians and genetics researchers who prepare data analysis with an agent and want a reproducible result.
Good fit when
- A VCF file must become a pharmacogenetic report with dosing by genotype.
- A list of variants needs disease links from nine databases at once.
- An analysis result must be rerunnable from saved commands and checksums.
- Patient data must not go to the cloud and everything runs on your machine.
Not a fit when
- You need a clinical decision: reports are research hints and not a medical conclusion.
- You rely on the ClawBio MCP server: it is removed in 0.8.0.
- You need the full demo data: install from source, not from the package.
Example request
Run the demo `clawbio run pharmgx --demo` and explain step by step which genes and drugs ended up in the report.Limitations
ClawBio is a research tool and does not replace a doctor: pharmacogenetic cards with recommendations must be checked against clinical sources. Installation needs Python 3.12 or newer. Some analysis replays depend on external databases and services that must be reachable. The full demo data set lives in the source and the package trims it. The MCP server is removed from version 0.8.0, and the skills themselves are used instead. The GitHub API reports no detected license, so read the LICENSE file in the repository. The last commit dates from November 2026.
How to disable. For the plugin run `/plugin` and remove ClawBio, for the package `pip uninstall clawbio`. Delete copied `skills/` folders from `~/.agents/skills/` by hand.
FAQ
How do I install ClawBio?
Run `pip install clawbio` (Python 3.12 or newer) or `conda install -c bioconda clawbio`. For Claude Code: `/plugin marketplace add ClawBio/ClawBio`.
How do I attach the skills in Cursor or Codex?
Copy or symlink the needed folders from `skills/` into `~/.agents/skills/` or the project `.agents/skills/`; no server is needed.
How do I use it as a library?
Import `from clawbio import run_skill, list_skills` and call `run_skill("pharmgx", demo=True)`.
What about the MCP server?
The `clawbio mcp` command was deprecated in 0.7.0 and removed in 0.8.0. The migration is described at docs.clawbio.ai.
Where does data come from, does it go to the cloud?
Analysis runs locally by default. Some skills call public databases, for example when searching variant associations.
Discussion
No questions or reviews yet.
Related
PluginsAround 200 skills for Claude Code on bioinformatics, chemistry and biostatistics, from RNA-seq analysis to molecular docking
17 skills for computational biology: RNA-seq, single-cell, protein structure, docking, literature and figures
A Claude Code plugin and MCP server: over 1000 scientific tools for biology, chemistry and medicine, data-check commands and routing skills
43 skills for healthcare and life sciences: genomics, clinical data, drug discovery, CDISC and claims analytics
StatsClaw assembles nine Claude Code agents to build statistical packages from a paper or formulas with isolated pipelines.
166 skills for scientific work: bioinformatics, cheminformatics, clinical data, geospatial analysis and 100+ databases
README in short
A Python package and a set of bioinformatics skills built on OpenClaw. The repository holds the `skills/catalog.json` catalog, a Claude Code plugin and docs at docs.clawbio.ai.