Ross ROSS = Recommend OSS · open-source software intelligence for agents

xuzhougeng/wisp-science

Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthropic models. observed · 2026-09-03

github.com/xuzhougeng/wisp-science · homepage · Rust · AGPL-3.0 (copyleft) observed · 2026-09-03

Health v2 · maintenance only

81/100

  • Activity 100
  • Release rhythm 100
  • Longevity 4

Flags: young

How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 1
  • age_days: 63
  • days_rel: 0
  • days_push: 0
  • n_releases_24m: 52

Full methodology

Adoption not part of the score

1087 stars · 113 forks observed · 2026-09-03

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Wisp Science is an open-source, local-first desktop AI research workbench that combines an LLM agent with persistent Python/R kernels, MCP-connected scientific databases, and SSH/WSL/GPU remote runtimes. It organizes literature, data, code, tool calls, and outputs into traceable, reproducible research projects while keeping data and credentials on the user's machine.

Use cases

  • run AI-assisted bioinformatics analyses like RNA-seq workflows locally
  • chat with an LLM agent that can execute Python and R code on my data
  • search PubMed and GEO and other scientific databases from an AI workbench
  • keep a reproducible record of analysis runs, figures, and decisions
  • run long compute jobs on remote SSH or WSL servers from a desktop app
  • use my own OpenAI or Anthropic API key with a local research assistant
  • organize papers, datasets, and code into one research project

When to choose

  • you want a local-first, privacy-conscious AI workbench where data and credentials stay on your machine
  • you do scientific computing in Python or R and want persistent, isolated kernels per conversation
  • you need MCP-based access to scientific databases like PubMed and GEO
  • you want reproducible research trails with approval gates and project-scoped artifacts
  • you need to orchestrate compute across laptop, WSL, and SSH/GPU servers

When to avoid

  • you need a hosted, zero-setup AI assistant that works without your own model API keys
  • you only need a plain chatbot without code execution or scientific tooling
  • you require a web-based multi-user platform rather than a desktop application
  • your workflow depends on cloud-native collaboration and automatic sync out of the box

Facets

application · maturity active

agent-framework mcp llm-inference chat-interface developer-tools sdk workflow-automation artificial-intelligence bioinformatics data-science large-language-models developer-tools self-hosted windows cli cross-platform local-first ai4science scientific-computing reproducible-research research-assistant bioinformatics-tools python-r-kernels ssh-remote-compute byom agent-skills ai-agents macos linux desktop tauri

5 sources

Member repositories

RepositoryRoleHealth v2
xuzhougeng/wisp-sciencemain81

For agents

markdown · JSON · MCP: product_card(name="xuzhougeng/wisp-science")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem