README
Sponsor
Atlas Cloud is a full-modal AI inference platform that gives developers a single AI API to access video generation, image generation, and LLM APIs. Instead of managing multiple vendor integrations, you connect once and get unified access to 300+ curated models across all modalities. Check out Atlas Cloud's new coding plan promotion for more budget-friendly API access: https://www.atlascloud.ai/console/coding-plan
Second Brain
Second Brain is a local-first AI runtime for your machine.
It indexes your files, remembers durable context, searches the web, runs tools, schedules cron jobs, sends Telegram updates, and lets agents extend the system while it is running. It is not a fixed chatbot wrapped around a folder search. It is a programmable conversation runtime with memory, retrieval, automation, and live plugin authoring built in.
The most important architectural shift is the conversation layer. Second Brain now routes conversations through a robust state machine: participants take actions, turns move between actors, phases suspend and resume multi-step flows, and frontends submit actions instead of owning conversation logic. Commands and frontends are plugins too, so the system can grow new user interfaces and slash-command workflows the same way it grows tools, tasks, and services.
What It Can Do
- Index documents, code, PDFs, slides, spreadsheets, archives, images, audio, and video.
- Search local files by keyword, semantics, or hybrid ranking.
- Answer from your own corpus with citations and exact file reads.
- Keep durable memory in
memory.md. - Store and resume conversation history in SQLite.
- Search the public web when local knowledge is not enough.
- Run path-driven indexing tasks and event-driven background jobs.