open-science
Open Science Desktop — local-first, model-agnostic AI research workbench for macOS, Windows & Linux. Open-source Claude Science desktop alternative built on Tauri + MCP + agent skills.
No outside reports on file
This tool runs on your machine with broad access to your files and can send data to external AI services, so only use it with research you're comfortable sharing and only install connectors from trusted sources.
Based on public adoption and transparency signals as of Oct 30, 2026. Our opinion, not a code audit or security guarantee.
What it can access
- Local filesNo local folder access notedLOW
- Command executionNo shell command runner notedLOW
- Network reachConnects to outside service APINOTED
- CredentialsNo API token requiredLOW
- Package sourceGitHub (ai4s-research/open-science)PUBLIC
Inferred from this listing’s public package details and description. Check the project’s source before installing.
Install
No npm or PyPI package is listed. Install from source: github.com/ai4s-research/open-science ↗
Everything known
We found nothing about this project in the sources we read (last checked 2026-10-30). That does not mean it has no problems.
0 incidents · 0 advisories · 0 news articles · 0 community threads · checked 2026-10-30
Popularity grade: C (from public signals such as stars and downloads, not a code review). Graded 2026-10-30.
Sources we read: OSV, GitHub Advisory Database, CISA Known Exploited Vulnerabilities, 44 security news feeds and our incident library. Last checked 2026-10-30.
About Open Science
Open Science Desktop is a workspace tool that helps researchers run AI experiments locally on their computer. It connects to AI models (including Claude) and lets you organize research workflows using MCP connectors and automated agent skills.
Who it’s for
Researchers and scientists who want to run AI-assisted work on their own machines without sending data to cloud services. It works on macOS, Windows, and Linux.
Alternatives
- Claude Desktop with MCP servers: Official Claude app with direct MCP integration for simpler setups
- Jupyter Notebook / JupyterLab: Local notebooks for AI-assisted research coding
- Cursor or VS Code with AI extensions: Code editors with built-in AI assistance for development workflows
Setup
Check these steps against the project’s own instructions before running anything.
Follow the project's README at https://github.com/ai4s-research/open-science (opens external site) for installation and configuration steps specific to your operating system.
Why this grade
Signal-based grade (3/7 points): well starred on GitHub, public source code. This reflects adoption and transparency, not a code review.
Popularity and upkeep
- GitHub stars
- 2.1k
- Listing last refreshed
- 2026-10-30
These are public popularity signals, not a maintenance check. Look at the repository’s commit history and open issues.
Risks to watch for
- Local file access: The tool can read and write files on your computer, so it could modify or delete research data if misused
- Model integration: It connects to AI models (like Claude), which means your research content gets sent to those services depending on your configuration
- Malicious MCP connectors: Third-party MCP connectors you install could be compromised and secretly steal your research or run unwanted commands
- Agent automation risks: Automated agent skills might perform unexpected actions if their instructions are unclear or conflicting
- Unvetted dependencies: As an open-source Tauri app, it relies on many libraries that could contain security flaws
How to lower the risk
Least-privilege permissions
- Only grant file access to specific research directories, not your entire home folder
- Use a separate user account or sandbox if handling very sensitive data
Network isolation
- Review which AI services (Claude, etc.) this connects to before setup
- Consider running it on an isolated network segment if possible
Monitoring
- Check what files and network requests the app makes using system tools
- Review logs for unexpected behavior before and after installing connectors
Red flags to watch for
- MCP connectors from unknown sources or with unclear descriptions
- Requests to run agent skills on sensitive files without clear explanation
- Unusual network traffic or file modifications after updates
Questions
How do I install Open Science?
Open Science does not publish a standalone npm or PyPI package. Clone or follow the setup instructions in its source repository at github.com/ai4s-research/open-science.
Are there security reports about Open Science?
No reports found for this package in our research (last checked 2026-10-30). Open Science currently holds Grade C (Moderate risk signals) based on public adoption and transparency signals. Key risk areas to watch include local file access, model integration, malicious mcp connectors.
Is Open Science actively maintained?
Open Science has 2.1k GitHub stars. Our directory data was last refreshed on 2026-10-30. Check recent commit activity, release tags, and open issues at github.com/ai4s-research/open-science before pinning it in production.
What are alternatives to Open Science?
Notable alternatives to Open Science include Claude Desktop with MCP servers (Official Claude app with direct MCP integration for simpler setups), Jupyter Notebook / JupyterLab (Local notebooks for AI-assisted research coding), Cursor or VS Code with AI extensions (Code editors with built-in AI assistance for development workflows). You can also compare Open Science side by side with other MCP servers in the Related tools section on this page.
README badge and team notes
Maintain or use Open Science? Show its live grade in your GitHub README, docs site, or run a check from the terminal before you install.
GitHub README Markdown:
[](https://agentarmorhq.com/skills/ai4s-research-open-science)HTML embed:
<a href="https://agentarmorhq.com/skills/ai4s-research-open-science"><img src="https://agentarmorhq.com/api/badge/ai4s-research-open-science" alt="AgentArmorHQ Risk Grade for Open Science" height="20" /></a>Terminal check:
curl -fsSL https://agentarmorhq.com/api/cli | node --input-type=module - check "ai4s-research-open-science"Share with your security lead
Need sign-off to use open-science? Copy this note for Slack, a ticket or a pull request.
### AgentArmorHQ Security Review Memo: open-science
- **Listing URL:** https://agentarmorhq.com/skills/ai4s-research-open-science
- **Type:** App connector
- **Risk Grade:** Grade C (Moderate risk signals)
- **Outside Security Advisories on File:** 0
- **Public Adoption:** 2.1k GitHub stars
#### Permission & Blast-Radius Signals
- **Local files:** No local folder access noted
- **Command execution:** No shell command runner noted
- **Network reach:** Connects to outside service API
- **Credentials:** No API token required
- **Package source:** GitHub (ai4s-research/open-science)
#### Recommended Hardened Setup
- **Source Repo:** https://github.com/ai4s-research/open-science
- **Controls:** Pin an exact version in `.mcp.json`, scope access to a single workspace folder, and use a least-privilege read-only token where supported.
_Based on public adoption and transparency signals as of Oct 30, 2026. Our opinion, not a code audit or security guarantee._Key facts
- Type
- App connector
- Publisher
- ai4s-research
- Source code
- github.com/ai4s-research/open-science
- Est. context load
- ~1.1k tokens / turn (~6 tool schemas, 0.53% of 200k context)
- Popularity
- 2.1k GitHub stars
- Last updated
- 2026-10-30
- Last graded
- 2026-10-30
Community
Discussions for Open Science
Ask questions or share configuration tips. Governed by our Community Rules & Terms.
No posts match this filter yet.
Be the first to ask a question or share a working tip.
Check before you install, in Claude
Your assistant answers Block, Review or No known issues for any package before it installs it.
Get this check inside Claude