org2AI/ORG2

System of record for how agents build software: Built-in rust harness & 20+ CLIs.

ORG‑2 – A record‑keeping hub for AI coding agents

What it is – ORG‑2 is a desktop application (built with Rust + Tauri) that acts as a system of record for any AI‑driven coding agent you run. It captures the whole “trajectory” of an agent session – the prompts you gave, the model’s understanding, tool calls, file edits, command output and the final Git changes – and stores it locally (under 100 MiB). The recorded sessions can be replayed like a video, commented on, and traced back to the exact decision that produced each line of code.

Why it matters – Traditional development tools only track the end result (commits, tickets, metrics). When agents generate code at speed, the reasoning behind a change disappears. ORG‑2 preserves that reasoning, giving teams an “AI‑blame” view that links every line of code to the agent run that created it. This helps with debugging, compliance, and aligning autonomous agents with human goals.

Core capabilities (as described in the README)

Feature What you get
Built‑in Rust harness Run fast, token‑efficient native agents using your own API keys and subscriptions.
Session manager Import and manage sessions from 20+ supported agent CLIs (Cursor, Claude, Codex, Copilot, etc.) and other tools, all in one UI.
Team collaboration Share sessions across devices, comment inline on the replay timeline, and review trajectories instead of just PR diffs.
Replay as video Synchronized view of messages, tool calls, file edits and command output – you can scrub forward/backward through a session.
AI‑blame Trace any line of code back to the originating agent session, tool call and decision, supplementing traditional Git blame.
Work diary / activity timeline Visual daily summary of time spent, tasks, and agent activity.
Full dev workspace Integrated terminal, Git view, pull‑request reviewer and source‑control actions without leaving ORG‑2.
Design Mode Embedded WebKit browser lets you inspect a live page, select an element and send its exact DOM/context to an agent for a fix.
Cross‑session memory & shared state Agents can read/write a common workspace state, enabling knowledge sharing across runs.
Resource‑aware execution The runtime can pause or throttle agents based on CPU, RAM or human‑attention constraints.
Scheduling & auto‑start Set agents to run overnight or continue work when you’re away.
Org‑level alignment (WIP) UI for linking sessions to issues/projects, exposing accountability at the organization level.
Self‑hosted Supabase collaboration (WIP) Optional backend for group issue workflows and persistent session storage.

Supported agents (selected)

  • Built‑in Rust harness (native)
  • Cursor CLI
  • Claude Code
  • OpenAI Codex
  • GitHub Copilot CLI
  • DeepSeek Harness
  • Kimi Code CLI
  • Aider
  • Goose
  • Amp
  • Grok CLI
  • Devin
  • Hermes
  • OpenClaw
  • Codebuff
  • Qwen Code
  • Mimo Code
  • Continue
  • Droid
  • Mistral Vibe
  • Autohand
  • OMP
  • Pi
    (and many more – the UI lists >20 CLIs)

Getting started

  1. Download – Pre‑built binaries are provided for macOS (Apple Silicon), Windows (installer or MSI) and Linux (AppImage & DEB).
  2. Run – The app launches a local‑first GUI; no server is required.
  3. Add agents – Use the built‑in Rust harness or point ORG‑2 at any of the supported CLI agents.
  4. Start a session – Launch an agent from the UI, give it a task, and ORG‑2 records the full interaction.
  5. Review – Open the recorded session, scrub through the timeline, add comments, or use the “AI blame” view to see why a line changed.

For developers who want to build from source:

pnpm install
pnpm run download:sidecars   # pulls optional native helpers (agent‑browser, peekaboo)
pnpm run tauri:dev           # launches the app in development mode

The repository also includes standard contribution docs (CONTRIBUTING.md, CODE_OF_CONDUCT.md).

Community & support

  • Discord: https://discord.gg/tvWgAqhCzs – channels for usage help, announcements, feedback, and general chat.
  • WeChat: linked via an issue thread.
  • Open‑source development is coordinated through GitHub issues and pull requests.

License

ORG‑2 is released under the GNU Affero General Public License v3.0 or later (AGPL‑3.0‑or‑later), meaning any public deployment of a modified version must also share its source code.

TL;DR

ORG‑2 is a lightweight, cross‑platform desktop app that records, replays, and shares the full execution history of AI coding agents. It turns opaque AI‑generated code changes into a searchable, commentable video‑like timeline, giving teams the ability to audit, debug, and align autonomous agents with human intent.

Related

  • Project
    HoangNguyen0403/agent-skills-standard

    Agent Skills Standard (AGS) is an NPM‑distributed CLI that lets teams publish their coding‑standards, security rules, and SDLC workflow templates as tiny markdown “skills”. The CLI (`ags init` / `ags sync`) creates a `.skillsrc` config, downloads language‑specific skill packs (280 skills across 20+ stacks), and writes a hierarchical router (`AGENTS.md → _INDEX.md → SKILL.md`). AI coding assistants (Cursor, Claude Code, Copilot, Gemini, etc.) read this router and load only the relevant rules for the file they are editing, saving ~85 % of prompt tokens. An optional MCP server can enforce hooks, block secret‑leak edits, and provide runtime audit logs. The project is a genuine AI‑coding‑assistant utility, not a generic tutorial or list.

  • Project
    tony1223/better-agent-terminal

    Better Agent Terminal (BAT) is a cross‑platform desktop app (Windows/macOS/Linux) built with Tauri 2 and React. It lets you manage multiple project workspaces, each with split‑panel terminals, a file browser, Git viewer, snippet manager, and an embedded AI coding assistant (Claude Code or OpenAI Codex). The AI runs inside the app, offering streaming output, permission modes, session resume/fork, and detailed usage stats. Optional integration with the `cx` CLI provides lightweight semantic code navigation. Install via Homebrew, AppImage, or a one‑line script; you can also build from source with Rust and Node. Ideal for developers who want a single window to run commands and interact with LLMs safely and efficiently.

  • Project
    vercel-labs/agent-browser

    A native Rust CLI that lets AI agents (or scripts) control Chrome via simple commands, using accessibility‑based element refs (`@eN`). It supports opening pages, taking snapshots, clicking/filling, extracting text, screenshots, network mocking, tab management, batch execution, and an experimental WebMCP tool system.

  • Project
    vastsa/PI-Desktop

    PI‑Desktop is a cross‑platform desktop app that provides a dedicated, local‑first workspace for AI coding agents. You connect any OpenAI‑compatible, Anthropic, or local model, open a local project, and let the agent read, edit, run commands, and iterate under three controllable modes (Agent, Plan, Goal). The app stores everything locally, offers a permission layer, supports plugins, skills, sub‑agents, and MCP integrations, and runs on macOS, Windows, and Linux.

  • Project
    GCWing/OpenBitFun

    An open-source desktop workspace for AI agents featuring a Rust runtime and multi-device synchronization to manage complex, long-running AI workflows.