# ClawBio/ClawBio

🦖 ClawBio - The first bioinformatics-native AI agent skill library. Local-first. Reproducible. Open. Free.

Repository: https://github.com/ClawBio/ClawBio
Canonical: https://ross.abutalabs.com/products/clawbio
Homepage: https://clawbio.github.io/ClawBio/
Language: Python
License: NOASSERTION
License Family: other
Topics: ai-agents, bioinformatics, equity, genomics, local-first, openclaw, population-genetics, reproducibility
Last push: 2026-09-02T08:47:03+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 95, longevity 13
- inputs: {"age_days": 189, "days_push": 0, "days_rel": 35, "gap_med": 16.0, "n_releases_24m": 7}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1123, forks 258 (observed 2026-09-03T02:15:05.298015+00:00)

## What it is
ClawBio is a bioinformatics-native AI agent skill library offering 97 curated, reproducible skills for genomics tasks like pharmacogenomics, GWAS lookup, and variant annotation. It runs locally via a Python CLI/library, installs as a Claude Code plugin, and ships an MCP server so agents in Cursor, Zed, VS Code, and Claude Desktop can execute real analyses with demo data by default.

## Use cases
- run pharmacogenomics analysis on genetic data locally
- query GWAS and genomic databases from an AI agent
- annotate variants in a whole-genome sequence
- give Claude or Cursor reproducible bioinformatics skills via MCP
- reproduce bioinformatics analyses with checksummed commands
- compute polygenic risk scores and nutrigenomics advice
- run scRNA-seq and population genetics workflows with demo data

## When to choose
- you want AI agents to execute real, reproducible bioinformatics pipelines locally
- you need privacy-focused genomics analysis with demo data by default
- you use Claude Code, Cursor, Zed, VS Code, or Claude Desktop and want genomics skills via MCP
- you want a curated, benchmarked library of bioinformatics skills instead of ad-hoc prompts

## When to avoid
- you need a hosted or cloud-native bioinformatics platform rather than local execution
- you require production clinical-grade genomic analysis with regulatory compliance
- you don't use AI coding agents and just need a traditional bioinformatics pipeline tool

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, mcp, cli, sdk, data-science
- domain: bioinformatics, artificial-intelligence, developer-tools
- platform: python, cli, cross-platform, self-hosted
- tags: bioinformatics-skills, genomics, mcp-server, reproducibility, local-first, pharmacogenomics, gwas, variant-annotation, claude-code-plugin, openclaw, ai-agents, command-line, docker

## Member repositories
- ClawBio/ClawBio (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:05.298015+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:41:26.809927+00:00, confidence not recorded.
  - readme: https://github.com/ClawBio/ClawBio (fetched 2026-09-03T02:15:05.298015+00:00, sha 6e3e5a985448)
  - homepage: https://clawbio.github.io/ClawBio/ (fetched 2026-08-29T12:45:26.676123+00:00, sha 9638fbea32c1)
  - site_page: https://docs.clawbio.ai (fetched 2026-08-29T12:45:26.679018+00:00, sha 6793d178f79f)
  - registry_pypi: https://pypi.org/pypi/clawbio/json (fetched 2026-08-29T12:45:26.681028+00:00, sha 504721670c8a)
- Data as of 2026-08-30T08:39:29.467469+00:00.