Why it is worth attention

It combines a capture-first daily feed with an LLM-powered wiki, agentic recall, and nightly consolidation – all running against your own API keys, with strict architecture and multi-tenant row-level security.

Who it is for

Knowledge workers and note-takers wanting a personal wiki
Developers building self-hosted, privacy-preserving knowledge systems
Privacy-conscious individuals who control their own LLM keys
Users who need offline-first mobile capture with sync

Use cases

Quickly capture random ideas via text or voice during the day
Automatically distill captured notes into a structured, interlinked wiki
Ask natural-language questions over your personal knowledge base
Run a nightly consolidation pass to keep the wiki coherent without manual effort

Strengths

Capture-first design with near-zero friction for inputting thoughts
BYOK (bring your own key) model – never shares credentials with the service
Offline-first Flutter app with stub mode for keyless exploration
Strict crate-layering law and CI-enforced architecture for maintainability

Considerations

Requires you to bring your own LLM API key (no built-in model)
Depends on PostgreSQL 16 and Redis 7 for full backend operation
Early-stage project – roadmap indicates current focus is clean open-source foundation, not broad provider coverage

README quick start

🧠 QCue

A capture-first “second brain” that turns fleeting thoughts into a living, linked knowledge wiki — powered by your own LLM keys.

English · 简体中文

[!NOTE] BYOK — bring your own key. QCue runs entirely against your own model-provider key. It never ships, brokers, or sees your credentials, and the whole product runs keyless and offline in stub mode so you can explore it in seconds.

QCue is an open-source knowledge, memory, and semantic-recall system. Dump fleeting ideas — typed or spoken — into a fast daily feed; a bring-your-own-key LLM continuously distills that stream into a persistent, interlinked Markdown wiki ([[wikilinks]], an index.md, a log.md) that you can browse, query by recall, and open in any Markdown editor. A nightly Auto-Dream pass keeps the knowledge base coherent.

📑 Table of contents

✨ What is QCue


🎙️ Capture-first	Type or speak. Ideas land in an append-only daily feed with zero friction — the cost of capturing a thought should be near zero.
🔗 LLM-wiki	A BYOK model distills your feed into a Karpathy-style Markdown wiki of entity / concept / source pages, woven together with `[[wikilinks]]`.
🔎 Agentic recall	Ask in natural language. The model generates its own search patterns over full-text search + curated memory and answers from your wiki.
🌙 Auto-Dream	A nightly consolidation pass compounds and tidies the wiki — cost-checked before any provider call, with edits proposed through a reversible approve gate.
🔌 BYOK & provider-agnostic	One ha

Description

Your second brain - capture your mind

Related repositories

Similar projects matched by category, topics, and programming language.

l0ng-ai

Featured

tty7

tty7 is a fast, GPU-accelerated terminal workbench with persistent sessions, built-in input tools, SSH support, and coding agent awareness.

Developer ToolsCLI & Terminal

359

steelbrain

Featured

reims-vgpu

reims-vgpu is an experimental virtual GPU for macOS guests that uses QEMU to decode the guest's GPU command stream and execute it through Metal or Vulkan, leveraging the built-in AppleParavirtGPU driver without requiring custom kexts.

Desktop Apps

143

m-novotny

Featured

memguard-rs

A Rust library that provides secure memory handling primitives including zeroization on drop, memory locking, constant-time comparison, and compile-time guarded regions, with zero dependencies and no_std support.

Embedded & IoTSecurity

131