open-science
Run traceable, reproducible AI research workflows in a local desktop workbench.
- Stars
- 729
- Forks
- 84
- Updated
- Updated Jul 14, 2026
Run traceable, reproducible AI research workflows in a local desktop workbench.
Open Science Desktop combines exploration, literature review, experiment code, notebooks, figures, reports, runs, and paper writing in one macOS, Windows, and Linux application. Each artifact can be traced back to its inputs, code, environment, model output, and conversation, while a bundled OpenCode sidecar keeps models, skills, and MCP connectors replaceable. Workspace data stays local by default, although selected content may still be sent to the model provider configured for a task.
Resource types
Use cases
Operating systems
Ask several AI chat services at once and compare their answers side by side.
Collect, organize, and cite BibTeX and BibLaTeX research libraries
Runtime
Protocols & integrations
Capabilities
Public GitHub facts last synced Jul 14, 2026.
Collect topic-based web content and discuss it with a local or cloud AI reading companion.
Analyze documents locally with collaborating agents and a persistent personal knowledge graph.