Last updated: 2026-06-26
CatGo is developed by the Wanlu Li Lab at the University of California, San Diego. We designed CatGo to keep your work on your own device. This policy explains what the CatGo mobile app (iPhone and iPad) does and does not do with your data.
- We do not collect any personal data.
- We do not use analytics, advertising, or tracking of any kind.
- We have no user accounts and require no sign-in.
- Your files, SSH credentials, and AI model keys stay on your device.
- Structure files and edits you open or create are processed locally on the device. They are not uploaded to us.
- SSH connection details and passwords/keys you save (to connect to your own HPC clusters) are stored locally using the device's secure storage. They are sent only to the server you specify when you initiate a connection.
- AI model API keys you enter are stored locally and are sent only to the AI provider you choose (for example Anthropic, OpenAI, Google, or DeepSeek) when you use the assistant. We never receive them.
CatGo only contacts external services in response to actions you take:
- Materials databases — when you search, CatGo queries public scientific databases such as the Materials Project, OPTIMADE providers, and PubChem. These requests contain only your search terms (e.g. a chemical formula), not personal information.
- Your HPC cluster — when you connect, CatGo talks directly to the SSH server you entered.
- Your chosen AI provider — when you use the assistant, your prompt is sent directly to that provider under your own API key, subject to their privacy policy.
CatGo has no backend server of its own that receives your data.
CatGo is a research and education tool and is not directed at children. We do not knowingly collect information from anyone, including children.
If this policy changes, we will update this page and the "Last updated" date above.
Questions: open an issue at https://github.com/Hello-QM/catgo-LRG or contact the Wanlu Li Lab at UC San Diego.