Features Built for AI-Usable Personal Knowledge
Legate Studio is designed from the ground up to make your personal knowledge accessible to AI tools. Every feature maps to a real workflow: capture, structure, connect, search, and expose to AI via MCP.
Get thoughts in before they disappear
The fastest path from a thought to your knowledge base — voice or text, from any device.
Record a voice memo on your phone or upload an audio file. Legate transcribes it, identifies the category it belongs to, extracts key concepts, and writes a structured note — automatically. No tags to apply, no folders to pick.
Paste in text from anywhere — meeting notes, article excerpts, raw ideas — and Legate processes it through the same pipeline. Transcription skipped; structuring, categorization, and knowledge extraction still run in full.
Every input goes through an AI pipeline: transcription (for audio), category assignment, concept extraction, and relationship mapping. The result is a structured knowledge entry, not a raw dump. Structure happens at capture time — before it's ever needed.
A library that's actually organized
Your knowledge lives in a searchable, browsable library — organized by category, browsable by date, and structured in a format AI tools can actually use.
Every note belongs to a category — determined by the AI at capture time. Categories are yours to name and define. Browse your library by category, date, or full-text search. Your entire knowledge base in one place.
Browse your knowledge chronologically. Daily journal view for what you captured today. Monthly and yearly rollups to see how your thinking has evolved. Your knowledge base as a record of your intellectual life.
Notes can carry task status. Track action items inside your knowledge base — without leaving the context where they appeared. Tasks live next to the knowledge that generated them.
See how your ideas connect
Every note is a node. Relationships between notes form edges. The result is a force-directed, interactive graph of your entire knowledge base.
Your knowledge visualized as a living map. Drag nodes, zoom in on clusters, click to open entries. The graph updates as you add new knowledge — patterns and connections emerge that you'd never spot in a list.
When new knowledge arrives, Legate identifies how it connects to what you already know. Relationships are built at capture time — not as a manual curation task. The graph fills itself in.
Related ideas form visible clusters. See which topics dominate your thinking, which ideas are isolated, and which concepts bridge multiple domains. The graph reveals the shape of what you know.
Find what you know, even when you don't know the words
Full-text search across your entire library. Semantic search finds related entries even when exact keywords don't match.
Search across the full text of all your notes — titles, content, tags, categories. Results appear as you type. Your entire knowledge base is instantly queryable.
Semantic search uses embeddings to find entries related by concept, not just by exact match. Ask a question in natural language and get results that are actually relevant — even when you used different words when you wrote them.
Narrow search results by category, date range, or both. Find all the notes from a specific project period, or everything you've captured about a particular topic. Filtering is fast and non-destructive.
Connect your knowledge to any AI workflow
Legate exposes your knowledge base via the Model Context Protocol (MCP) — so Claude, Cursor, and other MCP-compatible AI tools can query your memory directly.
Configure Legate as an MCP server in Claude Desktop. Your entire library becomes queryable from inside Claude — search, retrieve notes, create new entries, and browse categories without leaving your AI assistant.
Unlike chat history that resets, Legate's MCP integration gives AI tools access to a persistent, structured memory layer. Every conversation can start with relevant context — because your knowledge base remembers what the chat window doesn't.
MCP tools include: search library, get note by ID, list categories, create note, and more. AI agents can build workflows that read from and write to your knowledge base — Legate becomes the memory substrate of your AI stack.
Your data, your rules
Your knowledge lives in infrastructure you control. No data training on your notes. No lock-in. No surprises.
Each user gets their own SQLite database on a persistent Fly.io volume. Your data doesn't share infrastructure with other users. Export it, inspect it, migrate it — it's a file you can hold in your hands.
We don't read your notes, train models on them, or use your content to improve anything other than your own experience. Your knowledge base is private to you. Full stop.
Your knowledge base is a SQLite file. Download it directly from your account settings. No conversion required, no export wizard. Your notes are markdown text inside a standard database — readable by any SQLite tool.
Share what you choose, keep the rest private
Publish individual notes as clean public pages. Build a public profile to share your thinking with the world — while the rest of your library stays private.
Mark any note as public and it gets a clean, shareable URL at legate.studio/pub/you/note-slug. The page renders your markdown with proper typography and syntax highlighting — no styling work required.
Your published notes live on a profile page at legate.studio/pub/username. Set a display name, bio, and accent color. Your public profile shows only what you've chosen to share — the rest stays private.
Every profile gets an RSS feed at /pub/username/feed.xml. Readers can subscribe to your published notes using any RSS reader. A global feed at /feed.xml aggregates published notes from all users.
Start building your AI-usable knowledge base
14-day free trial. No credit card required. Full access from day one.
Start Free Trial