# CLI Manager (slim)

> CLI Manager is an open-source macOS desktop app that runs Claude Code, Codex CLI, Gemini CLI and shells side by side, with Git worktrees as workspaces. An opt-in local REST and MCP API lets another agent open and drive sessions.

- Full: https://www.anchorterminal.com/tools/cli-manager.md (~6,650 tokens) · this version ~1,480 tokens · JSON https://www.anchorterminal.com/tools/cli-manager.json · canonical https://www.anchorterminal.com/tools/cli-manager
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-15
- Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on

**B · 65.8/100 · rank #329 of 1084 · #9 in Agent harnesses · not agent-ready · confidence medium**


Assessment: An agent can open visible terminal sessions, type into them, wait and read the screen through 12 typed MCP tools or a documented REST API, and the user can take any session back. The API reaches every open session once enabled, the app runs only on macOS, and none of the last 35 CI runs on the main branch passed.

## Facts

- Kind: Agent harness · vendor: solhun (Jeong Kyeonghun) · category: Agent harnesses · legal entity: solhun (Jeong Kyeonghun) · provenance 50/100
- Local only (Streamable HTTP)
- Auth: API key · pricing: Free · x402: no · licence: MIT
- Probe metrics: not measured yet (probes haven't run)
- Surface listed: The macOS app v1.13.1 and its AI Control API (REST and MCP on 127.0.0.1), read from the repository, the docs page and the release notes on 14 October 2026. Nothing was installed or run
- Interfaces: Desktop app for people, plus the opt-in AI Control API. REST under `/v1` (health, workspaces, templates, sessions, output, input, wait, focus, rename, release, delete) and a stateless MCP endpoint at `/mcp` with 12 tools. An agent skill, `climanager-session`, wraps the REST API in a Node script
- Agents run: Claude Code, Codex CLI, Gemini CLI or any shell command, started from user templates. CLI Manager does not run a model loop of its own
- Approvals: Text refused while a question is on screen unless `force: true`. Disconnect AI and release hand a session back. No read-only mode for the API
- Sandbox: None. Sessions are ordinary login shells on the host with the user's rights
- Network: API bound to 127.0.0.1 with Host and Origin checks. Outbound traffic found in source is the update check against GitHub Releases and the user's `gh` CLI
- Git: Worktrees become workspaces, with stage, commit, push and pull requests through `gh` in the app. These are not exposed through the API
- Telemetry: None found in the source. The site FAQ says no data is sent externally
- Agent hooks: Optional Claude Code hooks and statusLine and Codex notify integration, off by default, which wraps rather than replaces existing commands
- Platforms: macOS Apple Silicon and Intel, signed and notarised DMGs. No Windows or Linux builds published
- Releases in 90 days: 12 (v1.7.0 on 17 August to v1.13.1 on 13 October 2026)
- Product Hunt: Second launch on 13 October 2026, ranked 17th that day. Product Hunt dates the first launch to 2025, and the site cites #10 Product of the Day

- Scores: Reliability 68, Performance pending, Schema & documentation 87, Agent ergonomics 63, Security & auth 58, Payments & pricing 60, Task success pending, Maintenance & community 62, Transparency & trust 54 · total over the 7 assessed categories
- Why: Reliability, Read as local software. · Schema & documentation, The MCP server lists 12 tools, each with a typed JSON Schema input and required fields marked. · Agent ergonomics, 12 compact tools (15). · Security & auth, Harness reading of the framework checklist, as for the other harnesses. · Payments & pricing, No payment protocol (0). · Maintenance & community, v1.13.1 on 13 October 2026 (30). · Transparency & trust, MIT in LICENSE and as detected by GitHub, while package.json still declares ISC (27).
- Sources: 22, open questions: 6, both in the full twin
- Capabilities: agent.multi-agent
- JSON: https://www.anchorterminal.com/api/v1/tools/cli-manager.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/cli-manager.svg` or a link to https://www.anchorterminal.com/tools/cli-manager from a page on solhun.com or one of its subdomains, or the README of github.com/woorichicken/CLI_manager, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Ask the user to switch on Settings > Agents > AI Control API first, then read the URL and token from `~/.climanager/control-api.json`
2. On `409 awaiting_input`, read the screen and answer with `keys` such as ["down","enter"]. Leave folder-trust and permission choices to the user
3. Stop using a session on `409 disconnected`. The user took it back
4. Treat `timedOut: true` from `wait` as still running, not as a failure, and wait again
5. Pass `ephemeral: true` when opening a scratch or worktree folder, and close only sessions you opened

## Connect

```bash
Download cli-manager-<version>-arm64.dmg or -x64.dmg from https://github.com/woorichicken/CLI_manager/releases/latest
```

```bash
curl -s -H "Authorization: Bearer $TOKEN" http://127.0.0.1:47821/v1/templates
```

```bash
claude mcp add --scope user --transport http cli-manager http://127.0.0.1:47821/mcp --header "Authorization: Bearer <token>"
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| OpenAI Agents SDK | AA | 86.2 | agent.multi-agent | https://www.anchorterminal.com/tools/openai-agents-sdk.min.md |
| Pydantic AI | A | 83.7 | agent.multi-agent | https://www.anchorterminal.com/tools/pydantic-ai.min.md |
| Microsoft Agent Framework | A | 82.2 | agent.multi-agent | https://www.anchorterminal.com/tools/microsoft-agent-framework.min.md |
| Docker Agent | BB | 76.5 | agent.multi-agent | https://www.anchorterminal.com/tools/docker-agent.min.md |
| Kilo Code CLI | BB | 75.4 | agent.multi-agent | https://www.anchorterminal.com/tools/kilo-code-cli.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)