# google-deepmind/science-skills

GDM Science Skills to speed up agentic scientific workflows with better grounding and higher token efficiency. Integrate insights from AlphaGenome, AFDB, UniProt and 30+ other databases and tools.

Repository: https://github.com/google-deepmind/science-skills
Canonical: https://ross.abutalabs.com/products/science-skills
Homepage: https://antigravity.google/use-cases/science
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-07-07T15:05:57+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 91, release rhythm 91, longevity 8
- inputs: {"age_days": 112, "days_push": 57, "days_rel": 63, "gap_med": 9.0, "n_releases_24m": 5}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2772, forks 309 (observed 2026-08-28T04:07:18.860051+00:00)

## What it is
A collection of agent skills from Google DeepMind that extend AI coding agents with structured instructions, scripts, and references for scientific research tasks. It covers genomics, structural biology, cheminformatics, and literature search, integrating data from AlphaGenome, AFDB, UniProt, and 30+ other databases and tools.

## Use cases
- run genomics analysis tasks through an AI agent
- query protein structure data from AFDB and UniProt
- perform cheminformatics workflows with an agent
- search scientific literature via OpenAlex from an agent
- speed up agentic scientific workflows with better grounding
- reduce token usage when agents do scientific research
- install curated science skills into Google Antigravity

## When to choose
- you use an agent platform like Google Antigravity or Gemini CLI for scientific work
- you need grounded access to bioinformatics and chemistry databases from an agent
- you want prebuilt, token-efficient skill instructions instead of writing your own

## When to avoid
- your work is unrelated to scientific research domains
- you need a standalone library with a programmatic API rather than agent skill files
- your agent runtime does not support the skills format

## Facets
- artifact type: plugin
- maturity: active
- function: agent-framework, sdk, developer-tools
- domain: artificial-intelligence, bioinformatics, data-science, education
- platform: python, cross-platform, cli
- tags: agent-skills, scientific-research, genomics, structural-biology, cheminformatics, literature-search, google-antigravity, token-efficiency, ai-agents

## Member repositories
- google-deepmind/science-skills (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:18.860051+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 2026-08-30T08:16:37.803392+00:00, confidence not recorded.
  - readme: https://github.com/google-deepmind/science-skills (fetched 2026-08-28T04:07:18.860051+00:00, sha 2c82fcfad2c1)
  - homepage: https://antigravity.google/use-cases/science (fetched 2026-08-29T09:56:03.040652+00:00, sha 4ab2f8af541a)
- Data as of 2026-08-30T08:39:29.467469+00:00.