GeminiLight/MindOS

MindOS is a Human-AI Collaborative Mind System, where human thinks and agents act. Globally sync your mind for all agents: transparent, controllable, and evolving symbiotically.

What it solves

MindOS addresses the problem of "memory silos" where context is scattered across different AI tools and chat sessions. It provides a local-first knowledge base that acts as a shared "second brain" for both the user and multiple AI agents, ensuring that personal profiles, project memories, and standard operating procedures (SOPs) are durable and accessible across different platforms.

How it works

MindOS stores knowledge as local plain-text Markdown files for privacy and ownership. It uses a Model Context Protocol (MCP) server and packaged "Skills" to bridge this local knowledge base to supported AI agents (such as Claude Code, Cursor, and Gemini CLI). This allows agents to read from and write to the shared context. The system includes a GUI workbench for humans to organize knowledge, a CLI for management, and a web clipper for capturing information from the web.

Who it’s for

It is designed for power users who use multiple AI agents and want a centralized, transparent, and controllable way to manage the context they provide to these agents, as well as a vontade to turn their interactions with AI into reusable instructions.

Highlights

  • Local-First Architecture: Knowledge is kept in local plain text, avoiding black-box memory systems.
  • MCP & Skills Bridge: Enables seamless integration with a wide variety of external AI agents via standardized protocols.
  • Human-AI Symbiosis: Allows users to distill reviewed conversations and corrections into reusable SOPs and Skills.
  • Comprehensive Tooling: Includes a desktop app, a browser extension for web clipping, and a YAML-based workflow orchestration engine.
  • Agent Inspector: Provides a filterable timeline of agent operation logs to audit tool calls and reasoning.

Related

  • Project
    rahilp/second-brain-cloudflare

    Second Brain is a Cloudflare‑hosted personal + team memory system that stores notes, decisions, and project facts and makes them searchable by meaning. It integrates with Claude, ChatGPT, Cursor, Obsidian, Notion, calendars, etc., and provides a secure, self‑hosted backend (D1, Vectorize, Workers AI) plus a “Prompt Capsule” feature for deterministic LLM context.

  • Project
    AVIDS2/memorix

    Memorix is a Node‑JS CLI/library that adds a local, SQLite‑backed, searchable memory store to any Git project. It lets AI coding assistants (Claude Code, Copilot, Cursor, etc.) share project facts, code snapshots, design rationales and commit‑derived knowledge across sessions and tools via the Model‑Context‑Protocol. Features include autopilot worksets, observation/reasoning memory, Git‑memory ingestion, code‑graph indexing, orchestration of multi‑agent tasks, a web dashboard, and optional sync. Install with npm, run `memorix setup --agent <agent>` and the agents can start reusing the shared memory.

  • Project
    doobidoo/mcp-memory-service

    A persistent, shared memory backend for AI agents that uses a knowledge graph and semantic search to maintain context across sessions.

  • Project
    dongshuyan/compass-skills

    A collection of reusable SKILL.md workflows that give AI agents persistent state, task tracking, and self-evolution capabilities for long-running projects.

  • Project
    oceanbase/powercontext

    A context management system that preserves work state, decisions, and progress across conversations to enable seamless hand-offs between humans and AI agents.