# aipoch/open-science

Open Science is an open-source, local-first, model-agnostic AI research workbench for reproducible scientific research on macOS, Windows, and Linux.

Repository: https://github.com/aipoch/open-science
Canonical: https://ross.abutalabs.com/products/open-science
Homepage: https://aipoch.com/open-science
Language: TypeScript
License: Apache-2.0
License Family: permissive
Topics: ai-for-science, local-first, reproducible-research, research-workbench, ai-research-tool, open-science, scientific-agent, scientific-research, ai-research, model-agnostic, research-software, scientific-ai, linux, macos, windows
Last push: 2026-08-26T22:33:04+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 99, longevity 4
- inputs: {"age_days": 61, "days_push": 7, "days_rel": 6, "gap_med": 1.0, "n_releases_24m": 45}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3182, forks 205 (observed 2026-08-28T04:07:47.823453+00:00)

## What it is
Open Science is an open-source, local-first, model-agnostic AI research workbench for reproducible scientific research. It combines a coordinator agent with specialist sub-agents and a reviewer, Python/R notebook execution, scientific data connectors, and traceable provenance in a single cross-platform desktop workspace.

## Use cases
- run an AI research assistant locally for reproducible scientific analysis
- automate literature review and hypothesis development with AI agents
- execute Python and R code in an AI-driven research workspace
- produce reports, tables, and figures with traceable provenance
- query scientific data sources from a unified workbench
- self-host an inspectable alternative to subscription AI research tools
- run data analysis and simulation across life sciences, chemistry, and physics

## When to choose
- you need reproducible, inspectable AI-assisted research with full provenance tracking
- you want a local-first, self-hosted workbench that works with any model provider
- you need Python and R execution plus scientific data connectors in one workspace
- you want an open-source alternative to fragmented research tooling

## When to avoid
- you need a lightweight CLI or library to embed in your own pipeline
- you require a fully hosted, zero-setup SaaS experience
- you need production-grade support or a long track record, as the project is young (v0.20.x)

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, llm-inference, rag, data-science, data-visualization, mcp, chat-interface, developer-tools
- domain: artificial-intelligence, data-science, education, self-hosted, cross-platform
- platform: windows, cross-platform, self-hosted
- tags: local-first, reproducible-research, scientific-research, model-agnostic, research-workbench, ai-agents, provenance, notebook-runtimes, open-science, research, science, macos, linux, desktop, typescript

## Member repositories
- aipoch/open-science (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:47.823453+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 2026-08-30T07:25:03.029470+00:00, confidence not recorded.
  - readme: https://github.com/aipoch/open-science (fetched 2026-08-28T04:07:47.823453+00:00, sha abea6bdaa3e8)
  - homepage: https://aipoch.com/open-science (fetched 2026-08-29T09:39:29.415014+00:00, sha d84abd308c51)
  - site_page: https://aipoch.com/guides (fetched 2026-08-29T09:39:29.446014+00:00, sha 49696ec4cc57)
  - site_page: https://aipoch.com/docs (fetched 2026-08-29T09:39:29.448336+00:00, sha 4fe5845f770c)
  - site_page: https://aipoch.com/agent-skills (fetched 2026-08-29T09:39:29.433397+00:00, sha bf21bdaf9190)
- Data as of 2026-08-30T08:39:29.467469+00:00.