Legal
Privacy Policy
Last updated: March 15, 2026 · Applies to Nemilia AI Workspace and the Nemilia Chrome Extension
The short version: Nemilia has no backend, no database, and no analytics pipeline. Your data never leaves your machine except to call the AI provider you explicitly configure. We have nothing to sell, nothing to share, and no server to store anything on.
1. Who We Are
Nemilia AI Workspace is developed and maintained by Luis Lopez. For questions about this privacy policy, contact luis@nemilia.com.
2. Nemilia AI Workspace (the App)
Nemilia AI Workspace is a single HTML file that runs entirely in your browser. It has no server, no backend, no database, and no cloud infrastructure of any kind.
What stays on your machine
- All documents, prompts, agents, workflows, and workflow results
- All API keys (encrypted with AES-256-GCM, PBKDF2 200,000 iterations, never written as plaintext)
- All agent memory entries
- All capture data from the Chrome extension
- All vector embeddings and RAG index data
- All chat history and conversation data
What leaves your browser (only when you initiate)
- LLM API calls — sent to your configured AI provider (OpenAI, Anthropic, Groq, etc.) when you run a prompt or workflow
- Web search API calls — sent to your configured search provider (Serper, Brave, Tavily, or SearXNG) when NEXUS or Chat web search runs
- Image generation API calls — sent to your configured image provider when VISION generates images
- CDN library loads — PDF.js, Transformers.js, WebLLM, and Chart.js load from public CDNs on first use
Telemetry
Nemilia has zero telemetry. No analytics, no error reporting, no usage beacons, no tracking pixels, no phone-home of any kind. The source is open and auditable — you can verify this yourself.
3. Nemilia Chrome Extension
The Nemilia Chrome Extension ("Send to Workspace") is a companion tool that captures web content and delivers it to the Nemilia AI Workspace application running in your browser.
What the extension accesses
- Page content — the extension reads the text content and images of pages you explicitly choose to capture via the right-click menu or popup. It does not read pages automatically or in the background.
- Active tab — accessed only when you trigger a capture action. Used to extract text, capture a screenshot, and retrieve the page title and URL.
- Local storage — used to temporarily queue captured items if Nemilia is not open. Items are delivered to Nemilia the next time it opens and are never sent to any external server.
What the extension does NOT do
- Does not send captured content to any external server
- Does not collect personal information
- Does not track browsing history
- Does not run on pages without explicit user action
- Does not share any data with third parties
- Does not use analytics or tracking of any kind
Where captured data goes
All captured content is delivered exclusively to the Nemilia AI Workspace tab open in your browser via the browser's local messaging API. It never leaves your machine.
4. Local AI Providers
When using Ollama, LM Studio, Jan, Llamafile, or WebLLM, all AI processing happens entirely on your device. No data is sent to any external server at any point.
5. Third-Party AI Providers
When you configure a cloud AI provider (OpenAI, Anthropic, Groq, etc.), your prompts and documents are sent to that provider according to their own privacy policies. Nemilia does not control, monitor, or log these communications. Review your chosen provider's privacy policy for details on how they handle your data.
6. Children's Privacy
Nemilia is not directed at children under the age of 13. We do not knowingly collect any information from children.
7. Changes to This Policy
If this policy changes materially, we will update the date at the top of this page. Continued use of Nemilia after any changes constitutes acceptance of the updated policy.
8. Contact
Questions about this privacy policy: luis@nemilia.com