^^ ^^ ^^ ^^ ^^
v0.1.0 — Open Source

Your thoughts,
locally powered.

A lean, local-first interface for capturing thoughts, recalling anything, and managing todos. All AI runs on your machine via Ollama. No cloud, no data harvesting.

min — Ctrl+Shift+Space
you > Meeting with David: discussed Q3 roadmap, need to follow up on API migration ^^ Saved 2 items (meeting, todo) — tags: david, Q3, roadmap, API ~~ Thought Link: “API refactor notes from last week” (84%) you > what did David and I discuss? ^^ You discussed the Q3 roadmap and the need to follow up on the API migration. [96% match] from your meeting note — saved 3 minutes ago you > /digest ^^ Today you captured 5 items: • Meeting with David about Q3 roadmap • 2 todos: API migration follow-up, code review for PR #42 • 1 thought about the new caching strategy
REACT 19 TAURI 2 RUST SQLITE OLLAMA
>_

Instant Capture

Global shortcut pops up a 416px chat window anywhere. Thoughts are stored in SQLite with auto-generated semantic tags.

Ctrl+Shift+Space
?~

Semantic Recall

Ask in plain language. Min embeds your query, searches by cosine similarity over stored vectors, and answers using only your data.

RAG_PIPELINE: ACTIVE
[]

Natural Todos

Create, list, complete, update and delete tasks with natural language. No forms, no checkboxes, just tell Min what to do.

TYPE: TODO
~~

Thought Links

When you save something, Min finds related notes via embedding similarity (threshold 0.7) and surfaces connections you missed.

COSINE_SIM: > 0.70
/d

Daily Digest

Type /digest for an AI-generated summary of today's thoughts, meetings, notes and todos — grouped by themes.

CMD: /DIGEST
^^

Quick Capture

Ctrl+Shift+T opens a minimal 480×72px floating input. Save a thought and it auto-closes. Zero interruption to your flow.

Ctrl+Shift+T

Local Models

Alibaba Qwen
Qwen 3.5 4B
Params 4 Billion
Size ~3.4 GB
RAM 6 GB
Nomic AI
Nomic Embed
Params 137 Million
Size 274 MB
RAM 4 GB
+ Any Ollama Model Configure in Settings

Five Agents, One Pipeline

Every message is classified into one of 5 intents and routed to a specialized agent. Each agent uses a dedicated skill prompt loaded from skills/*.md.

^^ Your Message
^^ Classification Agent
skills/classification.md — Routes intent to 1 of 5 agents
thought Save Decomposition save-decomposition.md

Splits input into storable items (thought, todo, note, meeting) with auto-generated tags. Preserves exact user wording.

decompose embed → store find similar → link
question Grounded Q&A question.md

Answers only from your stored data via RAG. Never uses training knowledge. Embeds query, searches vectors, augments prompt.

embed query semantic search RAG → respond
command Command Resolution command-decomposition.md

Resolves delete/update by matching natural language to stored document IDs. Asks for clarification when ambiguous.

search candidates resolve targets execute action
instruction Preference Store classification.md

Stores user behavior preferences and instructions that shape future responses and agent behavior.

extract preference store as pref
conversational Chat Agent conversational.md

Handles greetings, help, and casual chat. Explains Min's capabilities without executing storage actions.

stream response
^^
SQLite + Vector Embeddings Cosine similarity computed in pure Rust
thought todo note meeting

Keyboard-First

Toggle chat
Ctrl Shift Space
Quick capture
Ctrl Shift T
Send message
Enter
New line
Shift Enter
Close / hide
Esc
Custom shortcut
Settings
/digest AI summary of today's captured notes
/recall <query> Explicit semantic search with match scores
^^
Privacy Protocol

Data Sovereignty
by Design.

Most AI tools pipe your thoughts through remote servers. Min is different. All inference runs locally via Ollama. All data lives in a single SQLite database on your machine. We have no servers, no telemetry, no analytics. Your intellectual property stays yours.

0%
Data Collected
100%
Open Source
MIT
Licensed

Start Capturing

Free, open source, forever local.
Available for Windows, macOS, and Linux.