Orca CLI Skill for Claude: Drive Orca Worktrees and Terminals
orca-cli lets your coding agent drive the Orca desktop app: worktrees, terminals, agent handoffs and the built-in browser.
- Installs
- 357.1K
- on skills.sh
- Stars
- 112.4K
- on GitHub
- Updated
- Nov 2026
- last repo commit
orca-cli is a short launcher skill. It tells the agent which Orca command to run on your machine, then to load the full guide with orca skills get orca-cli. It triggers on phrases such as "Orca worktree", "child worktree", "spawn codex/claude in a worktree", "handoff" or "Orca browser", and it prefers Orca over raw git worktree or Computer Use when Orca state is involved.
You need the Orca app installed and running. The skill folder holds only this launcher, so what the agent does next comes from your installed Orca version, not from this page.
Good for
- Developers who run several coding agents in parallel Orca worktrees
- Orca users who want Claude to hand a task to Codex or another agent
- Anyone who wants an agent to read, wait on or send input to Orca terminals
Install orca-cli
Claude Code
- 1
Copy this prompt.
Install the orca-cli skill in Claude Code. Run this command: npx skills add https://github.com/stablyai/orca --skill orca-cli Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js.
From its listing on skills.sh. Source.
What's in the skill
When it loads
Operate Orca-managed worktrees, folder contexts, terminals, repos, automations, worktree comments, and Orca's embedded browser through the `orca` CLI. Use when the user says "$orca-cli", "Orca worktree", "child worktree", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "handoff" / "handover" / "give this to another agent", or "Orca browser". Prefer it over raw git worktree, ad hoc PTYs, or Computer Use when Orca state is involved. Use Computer Use only when a visible window needs GUI control that a CLI, filesystem, or API cannot do.
Sections in SKILL.md
- Resolve the CLI for this session
- Load the version-matched guide before running Orca commands
Key facts
- License
- MIT
- Installs on skills.sh
- 357,086 as of 2026-11-02
- GitHub stars
- 112,427 for the whole stablyai/orca repo
- skills.sh audits
- Gen Agent Trust Hub warn, Socket pass, Snyk pass
- What is inside
- SKILL.md only; bundles no other files
- Last commit to the repo
- 2026-11-01
What it includes
Chooses which Orca command to run, once per session
Uses ORCA_CLI_COMMAND if set, then orca-dev in a dev checkout, then orca-ide on Linux outside an Orca terminal, otherwise orca. On Linux a bare orca can start the GNOME screen reader.
Read ·
Pick the Orca executableLoads the guide that matches your installed Orca
Runs the chosen command to print the full usage guide before any other Orca command. If the command is unknown, the agent tells you to update Orca instead of guessing.
Try: “Use orca-cli to spawn Codex in a child worktree and hand off this task.”
Runs code ·
orca skills get orca-cliAsks Orca for machine-readable output
The agent prefers --json and uses --help for commands the guide does not cover.
Runs code ·
Prefer --json and --helpStarts Orca if a command says it is not running
Starts the app, then retries the command. It does this only for the not-running case.
Runs code ·
orca open --jsonRe-runs with approval if the sandbox blocks Orca
On runtime_access_denied the agent re-runs the command with escalated permissions and does not restart Orca.
Runs code ·
Stop on sandbox errors
From the official docs
Permissions
Medium risk
The agent can create worktrees and send input to terminals inside Orca. Approve commands you recognise, especially anything sent to a terminal.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
It can type into your Orca terminals
SKILL.md names reading, waiting on and sending to terminals. Check what it plans to send before you approve it.
It may ask for escalated permissions
If your sandbox blocks the connection to Orca, SKILL.md has the agent re-run the command with escalated permissions. Approve that only for the Orca command.
On Linux, bare orca can be a screen reader
SKILL.md warns that outside an Orca terminal the orca command normally starts GNOME Orca and speaks on your machine, so it uses orca-ide there.
Technical details
| Scopes |
|
|---|
Questions
What does the orca-cli skill do?
It teaches your agent to operate the Orca desktop app through its orca command: worktrees, folder contexts, terminals, repos, automations, worktree comments and the embedded browser. The skill itself is a launcher that loads the detailed guide from your Orca install.
Do I need Orca installed?
Yes. The skill only runs commands from the Orca app, and the stablyai/orca README lists downloads for macOS, Windows and Linux. Without Orca the agent has nothing to call.
How do I install orca-cli?
Use the install box at the top of this page, which has tabs for Claude Code, the Claude app and Codex. Use it in an agent that can run commands on your computer, since it calls the orca program.
Does it need an API key?
No key is named in SKILL.md.
orca-cli or orchestration?
Use orca-cli to hand work off when you expect nothing back, and for terminal control, shell commands, worktrees and the browser. Use the orchestration skill when you want a separate Orca worker agent to do a task and report back.
Is orca-cli safe?
It can create worktrees and send input to your Orca terminals, so treat it like any agent with shell access. It bundles no scripts, and SKILL.md tells the agent to ask for escalated permissions rather than work around a blocked sandbox. See the audit results under key facts.
Information
updated Nov 2, 2026
- Repository
- stablyai/orca
- Folder
skills/orca-cli
Sources
Docs, issue trackers and discussions this page is built from.
Show all 3Hide
- orca-cli SKILL.md · github.com
- stablyai/orca README · github.com
- skills.sh listing and audits · skills.sh