<- back to runpane.com

runpane cli

the cli that lets agents use pane

runpane installs Pane and gives agents the commands they need to create panes, read terminal output, and send input. It works on Windows, WSL, macOS, Linux, local machines, and remote hosts.

every os, first-class

Windows and WSL aren't afterthoughts. The same workflow runs on macOS, Linux, VMs, and remote hosts too.

npm, pnpm, and pypi

Start with npx, pnpm dlx, pipx, or install globally when agents need runpane on PATH.

agents can control pane

Any agent that runs shell commands can add repos, create panes, check terminal output, and send input.

built for agent loops

Your agents can use Pane as the workspace they manage. No hosted scheduler, no special SDK. Just terminal commands.

install

For a one-time install or update, use a runner. It grabs the latest version and doesn't leave anything behind:

npx --yes runpane@latest
pnpm dlx runpane@latest
pipx run runpane

For agents or scripts that call runpane repeatedly, install it globally:

npm i -g runpane
runpane setup

python -m pip install runpane
python -m runpane setup

why a cli instead of mcp

MCP is great for external tools. Agents can use MCP for GitHub, Linear, Jira, Slack, and anything else. Pane doesn't replace those.

But every agent already has a terminal. A CLI works in Codex, Claude Code, Aider, Goose, OpenCode, shell scripts, cron, and CI. No extra server to install, no per-agent setup.

It's lightweight by default. runpane agent-context prints a short summary. Agents only load full command details when they actually need them.

Agents can read it before installing. The full command list is public at /runpane-cli-contract.json.

The full workflow is open source. The Codex and Claude Code orchestrator skills handle the full issue-to-PR loop.

One CLI for both humans and agents. Installing, updating, diagnostics, creating panes, reading output, sending input. It's all the same runpane command.

the core loop commands

runpane doctor --json
runpane agent-context
runpane agent-context --command "panels wait" --json

runpane repos list --json
runpane repos add --path "$PWD" --yes --json
runpane agents doctor --agent codex --repo active --json

runpane panes create --repo active --name "issue-252" --agent codex --prompt "Plan this issue" --source agent --wait-ready --yes --json
runpane panes list --repo active --json
runpane panels list --pane <pane-id> --json
runpane panels wait --panel <panel-id> --for ready --json
runpane panels screen --panel <panel-id> --limit 80 --json
runpane panels output --panel <panel-id> --limit 200 --json
runpane panels submit --panel <panel-id> --text "Continue" --yes --json
runpane panels submit-composer --panel <panel-id> --strategy auto --yes --json

wsl and windows

If your agent runs inside WSL but Pane is installed on Windows, call the Windows wrapper through PowerShell:

powershell.exe -NoProfile -Command \
  'Set-Location $env:TEMP; runpane repos list --json'

where to go next

- Set up your first AI agent loop

- Read the runpane CLI reference

- See the full CLI command list

- Use the RunPane Orchestrator skill for Codex

- Use the RunPane Orchestrator skill for Claude Code

- Agent loop docs

frequently asked questions

or
Download*
Windows SmartScreen warningDirect downloads can show a SmartScreen warning while Pane is unsigned. Pane is fully open source, so you can audit the code and build from source yourself.1. Click More info2. Click Run anyway3. Continue the installerThe PowerShell install downloads the official release directly and avoids most browser download friction.npm global install
Paste that in PowerShell.MacLinux