Unterm

Unterm

The terminal AI agents can drive.

Download the latest release · 🪞 AtomGit 国内高速下载 (China Mirror) · Website · Docs

If Unterm is useful to you, a ⭐ on GitHub or AtomGit helps other people find it.

Cross-platform terminal (macOS / Linux / Windows) built on Unterm's native next-core terminal engine, with one design bet: the terminal itself is controllable from the outside by any AI agent over MCP. Claude Code, Codex, Gemini CLI, Cursor, Aider, your own scripts — they all get the same JSON-RPC surface (152 authenticated methods plus `auth.login`) to spawn shells, run commands, read pane state, capture screenshots, record sessions, and run the Agent Cockpit. (Theme, language and unterm.conf are not on MCP; they are set through Web Settings, unterm-cli, or the file.)

Since v0.68 the terminal is also something an orchestrator can govern rather than merely call: it publishes what it can do and how dangerous each capability is, works under leases that expire and cannot be replayed, keeps agents inside workspaces that cannot see each other, and can hand you an evidence bundle for a task that somebody who was not there can verify.

Since v0.55 the relationship runs both ways: agents drive the terminal from outside, and the terminal is an Agent Cockpit for the agents running inside it — live per-pane agent state, a waiting-first Inbox, fleets of N agents on one task in N isolated git worktrees, and a Review page to diff / merge / roll back what they produced.

The other 2026 terminals each pick a different side: Warp embeds AI inside a closed cloud (Oz), Ghostty stays out of your way and lets you bring your own tools, iTerm2 is Mac-only. Unterm picks the third side — terminal as MCP-controllable surface, deliberately keep AI *generation* out of the terminal, let external agents grip it through the API, and give the human one cockpit to run them all from.

Practical implications:

• Every Unterm window starts a local MCP server (line-delimited JSON-RPC over TCP) and a local HTTP settings server (Web Settings page) on auto-allocated ports. Both are auth-token gated, both are 127.0.0.1-only, no cloud round trip.
• Settings live in the browser, not the terminal. Cell-grid TUIs can't deliver modern form UX (no proper text inputs, no live preview, no color picker). The in-terminal ▼ menu holds quick actions and links out to the Web Settings page — configuration itself happens in the browser.
• 9 languages out of the box: en / 简体中文 / 繁體中文 / 日本語 / 한국어 / Deutsch / Français / Italiano / हिन्दी. Auto-detects from system locale, can be overridden in Web Settings or via unterm-cli lang set .
• Multi-instance discovery: every running Unterm process owns one NATO-named instance (alpha, bravo, charlie…) and writes its ports + auth token to ~/.unterm/instances/ .json. Agents that drive several windows at once enumerate that directory.
• Cross-platform parity is a correctness property: if a feature works on Windows but bails on macOS or Linux, that's a bug, not "not supported yet."
• Subtraction over decoration: no AI chat overlay inside the terminal and no cloud dependency for core operation. Proxy settings auto-detect the system by default and also support explicit HTTP/SOCKS overrides, node pools, rotation, and Clash/mihomo controllers. Finder integration on macOS uses the native Finder right-click extension and Services.

The GUI and terminal runtime now use Unterm's native next-core engine. The repository still carries selected upstream components and attribution where they remain dependencies, but WezTerm mux/window state is no longer the product kernel.

---

Agent Cockpit

Run Claude Code, Codex, Gemini CLI, or Aider in any pane and Unterm sees them — no configuration, no wrapper. Five pillars, all local:

• Agent state engine — every pane's agent and its state (working / waiting-for-you / idle / done), read from OSC progress + title signals, process fingerprints, and optional official hooks. Tab badges + a cross-window tally chip in the top bar.
• Inbox (Ctrl+Shift+A) — every agent that's waiting for you in one queue, longest-waiting first. Enter jumps to the pane; one keystroke later you've answered its prompt.
• Fleet — one task × N agents × N isolated git worktrees (../ .fleet/), one tab each. Same agent ×3 for throughput, or claude,codex,gemini for a bake-off.
• Review — agents get checkpointed before they touch a repo (dangling-commit snapshots; nothing touches your HEAD or index). A Web Review page shows per-member diffs with squash-merge (stops at staged — the commit stays yours), discard, and rollback. Since v0.57, Review also verifies each member (inferred or explicit validation command), ranks members by verification + change size, gates merge on a passing run, and can retry a failed member in its existing worktree.
• Everything scriptable — the cockpit itself is MCP + CLI: agent.status, cockpit.inbox, fleet.launch, review.merge… an orchestrating agent can run fleets and review diffs with no human in the chair.

unterm-cli agent status                                  # who's running where, in what state

unterm-cli agent inbox # who's waiting for you unterm-cli agent enable-hooks # exact state via official hooks (merge-only, backed up) unterm-cli fleet launch --agents claude,codex "fix the flaky auth test" unterm-cli review verify --fleet --member 1 # run the member's validation unterm-cli review list && unterm-cli review open # ranked diffs in the browser

Full docs: unterm.app/docs/agent-cockpit.

---

Install

Pre-built artifacts are published on GitHub Releases and mirrored on AtomGit (for fast direct downloads in mainland China):

• Global (GitHub Releases): https://github.com/zhitongblog/unterm/releases
• 中国大陆高速下载 (AtomGit 镜像 / G-Star): https://atomgit.com/zhitongblog/unterm/releases
• macOS (Universal arm64+x86_64): https://atomgit.com/zhitongblog/unterm/releases/download/ /Unterm-macos- .dmg
• Windows x64: https://atomgit.com/zhitongblog/unterm/releases/download/ /Unterm- -x64.msi · Unterm-windows-x64- .zip
• Windows ARM64: https://atomgit.com/zhitongblog/unterm/releases/download/ /Unterm- -arm64.msi · Unterm-windows-arm64- .zip
• Linux x86_64: https://atomgit.com/zhitongblog/unterm/releases/download/ /unterm- .deb · Unterm- -x86_64.AppImage
• Linux ARM64: https://atomgit.com/zhitongblog/unterm/releases/download/ /unterm- .arm64.deb · Unterm- -aarch64.AppImage

macOS

Double-click Unterm-macos- .dmg, then drag Unterm.app onto the Applications shortcut. The DMG is signed with a Developer ID and Apple- notarized, so Gatekeeper opens it on first launch without warnings.

Finder integration is bundled in the DMG. After the first launch, Finder's right-click menu can show Open in Unterm for folders and files; if macOS doesn't refresh the extension immediately, run Repair Finder Integration.app from the DMG once.

Linux (Debian / Ubuntu)

sudo apt install ./unterm-<version>.deb

unterm

Other distros — use the AppImage:

chmod +x Unterm-<version>-x86_64.AppImage

./Unterm- -x86_64.AppImage

Windows

Run the MSI installer; it places unterm.exe in Program Files\Unterm and creates a Start Menu shortcut.

---

What's new

• v0.68 — A terminal something else can govern. Unterm now hosts CLI agents as sessions rather than shelling out and waiting for an exit code (agent_session.*: what it said, what it asked to run, how it ended, with your own task ids carried through untouched). It can lease a browser from Unzoo and be leased *from* in turn — terminal.manifest publishes what this terminal can do and how dangerous each family is, taken from the same table the gateway refuses by. Approvals can finally be answered: Settings shows what an agent is waiting on, with "allow once / for this task / always". Workspaces are roots that cannot see each other, and a shell that cds out stops being inside. The audit trail is hash-chained, so an edit to it disagrees with the next line. unterm-cli provider | scope | artifact | evidence | system — 46 new MCP methods (149 total at the time; 152 today).
• v0.57 — Fleet verification loop + new brand mark. Review now verifies each fleet member automatically (Cargo / Go / npm / pnpm / yarn / Python / Maven / Gradle / .NET inferred, or your own command), ranks members by verification and change size, gates squash-merge on a passing run (audited force override), and retries failed members in their existing worktree without losing work — review.verify / fleet.retry over MCP + CLI. The sidebar gains repository-grouped navigation with always-on fuzzy search. Every logo surface moves to the new command-loop mark.
• v0.55 — Agent Cockpit. The terminal now sees the agents inside it: live per-pane state with tab badges and a cross-window tally, the waiting-first Agent Inbox (Ctrl+Shift+A), fleets running one task across N agents in N isolated worktrees, and a Review page with checkpoints, diffs, rollback, and squash-merge. 12 new MCP methods, 3 new CLI families.
• v0.54 — 2.8× faster cold start (~780ms → ~280ms) via five startup-path wins, and no more CPU core burned on Windows output floods (~91% → ~4%); MCP stays responsive mid-flood.
• v0.53 — Composer + Git panel. A prompt queue (Ctrl+Shift+J) that runs batched prompts into an agent pane with smart auto-advance through confirmation prompts, and a read-only Git status panel (Ctrl+Shift+G).
• v0.52 — More agents out of the box. Kimi Code CLI and Trae Agent join the baked manifest (7 agent CLIs total); reworked per-frame paint paths; steadier Windows clipboard and window sizing.

---

Documentation

The full Unterm docs live at https://unterm.app/docs/:

• Agent Cockpit — agent state engine, Inbox, Fleet, Review: run and supervise CLI agents from one terminal
• Agent integration — how to drive Unterm from Claude Code / Cursor / Aider / your own client
• Agent recipes — copy-paste patterns for common agent-drives-terminal workflows
• Product roadmap — what is being built now and what comes later
• [Product requirements](docs/product-requi