melandlabs/openloomi

OpenLoomi is an open-source AI coworker. It connects your work tools, understands what you’re working on, and tells you what needs your attention, why it matters, and what to do next. It’s your open-source Attention Agent.

OpenLoomi – an open‑source “attention agent” for your desktop

What it is

  • A native desktop application (Windows, macOS, Linux) that runs continuously in the background and acts as a personal AI coworker. It watches the apps you use, pulls in data from your email, chat, issue trackers, etc., and builds a searchable, auditable memory of people, projects, decisions and tasks.
  • The UI shows a small “Loomi” bubble that can surface reminders (e.g., 9 am to‑do list, overdue replies) or let you ask questions directly from the messaging apps you already use (Telegram, WhatsApp, iMessage, QQ, Lark/Feishu).

Key capabilities

Feature What it does
Attention Agent Always‑on desktop companion that pops up decision cards at the right moment without stealing focus.
Holistic Context / Memory Short‑, mid‑ and long‑term memory that automatically records facts from your tools; the memory is visible, auditable and stored locally with AES‑256 encryption.
Platform Connectors Background sync pulls commits, issues, emails and documents into a unified context graph; messaging‑app connectors let you chat with the AI inside the apps you already use.
Proactive Tasks Schedule recurring digests, reports or reminders that run automatically on your machine.
Security & Ease of Use Local‑first, no data leaves the device, open‑source, simple installer, works out‑of‑the‑box.
Agent‑agnostic integration Exposes its memory, connectors and loop engine as Skills and Plugins that can be used by any LLM‑powered agent runtime (Claude Code, Codex, OpenCode, Hermes, OpenClaw, etc.).

How you use it

  1. End‑user – download the installer from the GitHub releases page (DMG for macOS, EXE for Windows, .deb/.rpm for Linux) and run it. The app starts a local API and shows the Loomi bubble.
  2. Agent plugin – add the official OpenLoomi plugin to a supported agent (e.g., Claude Code or Codex CLI). The plugin turns the agent into a front‑end for the local OpenLoomi runtime, letting you issue commands like /openloomi:setup.
  3. Skill‑only runtimes – install the repository’s skill set with npx skills add … and run the openloomi or openloomi-setup skill. The skill checks that the desktop app is installed and talks to its local API.
  4. Developer – clone the repo, install Node 22+, pnpm 9+, Rust 1.75+ and run pnpm tauri:dev to build the Tauri‑based desktop client.

Why it matters

  • Open‑source & neutral – unlike many “AI coworker” products that lock you into a single vendor, OpenLoomi works with any LLM agent you prefer.
  • Context beyond code – it remembers not just repository changes but also emails, chat threads, decisions and project status, giving agents a richer, time‑aware knowledge base.
  • Local‑first privacy – all data stays on your machine, encrypted, with audit logs, so you retain full control.

Where to learn more

License – Apache 2.0 (permissive, allows commercial use and modification).

Related

  • Project
    beilusaiying/always-accompany

    A multi-purpose AI agent framework focused on context and attention mechanisms, providing layered memory and dynamic retrieval for long-term companionship, coding, and work.

  • Project
    NxcoreAI/EverRoom

    A local-first personal context workspace that organizes files, conversations, and memory into scoped "Context Rooms" to provide AI agents with grounded, traceable context.

  • Project
    getomnico/omni

    An open-source, self-hosted AI agent for companies that integrates with workplace tools to gather context, analyze data, and execute tasks across systems.

  • Project
    zeenie-ai/OpenCompany

    OpenCompany is an open‑source desktop app that lets you create autonomous AI “employees” – small teams of LLM‑driven agents – that can read email, write code, post to social media, handle payments, etc. You define a job, drag specialist agents onto a visual canvas, connect them to tools (Gmail, GitHub, Stripe, browsers, Android emulator, …), and the employee runs locally (or in Docker) using any LLM you provide. State, memory, and skills are editable, and the system works without a subscription.

  • Project
    tinyhumansai/openhuman

    OpenHuman is an open‑source Rust‑based AI‑agent harness offering a desktop app and a library. It runs hundreds of agents on a cheap VPS by compressing prompts, using a tiny tool‑search model, and keeping the core in a single compiled process. Benchmarks show it’s the fastest, cheapest, and most memory‑efficient among seven comparable harnesses. Features include built‑in memory, voice agents, real‑browser/desktop control, 119 OAuth integrations, 14 messaging channels, and support for local or BYOK models. Install via pre‑built binaries or a simple script; developers embed it with a few lines of Rust code.