The MCP-First Personal Knowledge Base for AI
Capture voice and text, structure it into searchable memory, and connect your knowledge to AI workflows. Not a chat app. A memory substrate you own and control.
Full access for 14 days. Then BYOK from $0.99/month or Managed at $9.99/month.
Capture → Structure → Connect to AI
Three steps from raw thought to AI-accessible memory. No manual tagging. No reorganizing. Just capture and go.
Voice memo, audio file, or text — Legate's AI transcribes and categorizes it automatically. Capture from your phone on the go, or paste text directly. No organizing required at capture time.
Every entry becomes a titled, categorized node in your personal knowledge graph. Connections form automatically. Search finds it semantically. Your library grows in a structure AI can actually use.
Connect Legate to Claude Desktop or any MCP-compatible AI tool. Your AI can now search, retrieve, and create notes in your knowledge base directly — no context copy-paste, no tab switching.
What MCP-first actually means
Most note apps were built before AI mattered. Legate Studio was built after — with AI integration as the primary interface, not an add-on.
MCP (Model Context Protocol) is the open standard that lets AI assistants reach outside their own context window and query external tools and data sources. When your AI has an MCP connection to your knowledge base, it's not working from memory or generic training data — it's working from your actual notes, in real time.
Most personal knowledge management tools treat AI as a feature layered on top: a chat window in the sidebar, a summarize button, an AI-powered search. MCP-first means the opposite. The MCP interface is the primary way the knowledge base gets used. The web app is where you capture and organize; the MCP server is where AI gets work done with your knowledge.
Practically: you connect Legate Studio to Claude Desktop by adding one config entry. From that point, your AI assistant can search your knowledge base, retrieve specific notes by category or keyword, and create new entries — all from inside the AI conversation, without you doing anything manually.
This changes what AI is useful for. An AI that knows your context — your projects, your decisions, your terminology — is a different tool than one that's starting from zero every session.
Features built for AI-usable personal knowledge
Every feature was designed around the question: can an AI actually use this? See full feature list →
Record voice memos or upload audio files from anywhere. AI transcribes, categorizes, and extracts structured notes automatically. No manual organizing at capture time — just speak and go.
A visual, force-directed map of how your ideas connect. New entries find their place automatically. See clusters form as your knowledge grows — patterns emerge that flat lists never reveal.
Search by meaning, not just keywords. Find notes based on what they're about, not just what words they contain. Available from the web app and as an MCP tool your AI can call directly.
Connect Legate to Claude Desktop and other MCP-compatible AI tools with a single config entry. Your AI can search, retrieve, and create notes in your knowledge base without leaving the conversation.
Publish individual notes as clean, readable public pages. Build a public profile with your own accent color and bio. Share specific insights while keeping the rest of your library private.
Every captured thought becomes a categorized, titled entry in your personal library. Browse by category, date, or project. Search finds what you need. Structured enough for AI to navigate.
Your data, your control
We built Legate Studio around a simple principle: your knowledge is yours. Not in a marketing-copy sense — in a technical, architectural sense.
Your knowledge lives in a GitHub repository you own. Full version history, portable, exportable. Clone it anytime — no export process, no proprietary format. Your notes are Markdown files.
We don't read your notes, train on your content, or store your knowledge beyond what's required to run the app. Your GitHub repo is private by default. We can't access it even if we wanted to.
Because your data is in a GitHub repo you control, leaving Legate Studio means nothing is lost. Clone the repo, delete the app — your knowledge is still there, readable by any Markdown editor.
How people use Legate Studio
Different workflows, same underlying pattern: capture → structure → connect to AI.
What it looks like when your note app and your AI share the same memory layer.
Solve the stateless AI problem with structured personal knowledge that persists across sessions.
From phone recording to searchable, AI-accessible knowledge entry in seconds.
What a personal knowledge base that's actually built for AI looks like under the hood.
Why connected notes produce better recall and better AI context than flat lists.
Give your AI assistant durable context that survives session boundaries.
Simple, transparent pricing
Start free. Upgrade when you're ready. No surprise fees, no lock-in. Full pricing details →
Full access to every feature. No credit card required.
- Full access to every feature
- Voice → note pipeline
- Library & knowledge graph
- MCP integration
- Public profiles & publishing
- No credit card required
Bring your own Gemini API key. Full platform at the lowest possible price.
- Everything in Free Trial
- Bring your own Gemini API key
- You pay Google directly — no markup
- Full MCP integration
- Your data, your keys
We handle the AI. No API keys to configure, works out of the box.
- Everything in BYOK
- Gemini Flash included
- No API keys to configure
- Full MCP integration
- Zero setup, zero hassle
Common questions
The short answers. See the full FAQ →
Build your AI memory layer today
14-day free trial. No credit card required. Full access from day one — voice capture, knowledge graph, MCP integration, everything.