Manage GeeLark from your terminal
Manage cloud phones, browser profiles, and supported automation tasks with command-line tools. Run everyday operations directly or make them part of your scripts.
Find API, CLI, and MCP connection options in GeeHub in the GeeLark client.
$ geelark-cli phone list --page 1 --page-size 10 # Sample output (illustrative) NAME GROUP STATUS Phone A Group 1 running Phone B Group 1 stopped Phone C Group 2 running
Cloud phones
Browser profiles
Make everyday GeeLark operations repeatable
Check your workspace from the terminal
Retrieve available cloud phone, browser profile, and resource information without switching between multiple interface views.
Put repeat operations into scripts
Use supported commands inside your own scripts to repeat routine operations with the logic and checks your workflow needs.
Work with structured results
Use JSON output for further processing, or choose a table when you want a readable view in the terminal.
Start with a few practical commands
List cloud phones
Retrieve a page of cloud phone profiles.
$ geelark-cli phone list --page 1 --page-size 10List browser profiles
Retrieve a page of browser profiles using the configured browser connection.
$ geelark-cli browser list --page 1 --page-size 10Choose a readable output format
Display cloud phone information in a terminal-friendly table.
$ geelark-cli phone list --format tableReview task history
Retrieve recent cloud phone automation task history.
$ geelark-cli phone automation task-history --size 10Commands require the appropriate authentication, permissions, and connection setup. See the official reference for current parameters and requirements.
Browse all commands →Control profiles, tasks, and resources
Cloud phone and browser commands can have different prerequisites. Check the official documentation for the command you want to use.
| Area | Supported examples |
|---|---|
| Cloud phones | List, create, start, stop, update, and inspect supported cloud phone profiles. |
| Browser profiles | List, create, edit, launch, and stop browser profiles. |
| Automation | Work with supported custom tasks and task-management commands. |
| Proxies | List, add, update, and check proxies. |
| Groups and tags | Organize supported resources by project. |
| Apps and files | Access documented cloud phone app, file, and Library operations. |
Use GeeLark commands in your scripts
Keep your own logic around supported GeeLark operations. Retrieve information, select the resources your script should use, run the relevant commands, and process the returned results.
- 01
Retrieve information
- 02
Apply your script logic
- 03
Run supported operations
- 04
Review results
✓ Reuse commands for recurring work.
✓ Process JSON output with your existing tools.
✓ Keep execution checks and error handling in your own scripts.
Scheduling, retries, and orchestration must be implemented by your script or chosen environment unless a specific GeeLark command provides them.
Choose CLI, API, or MCP
Find all three in GeeHub. Choose the interface that fits how you work.
| Interface | How you use it | Best for |
|---|---|---|
| CLI | Commands from your terminal or scripts | Best for repeatable command-line operations |
| API | Requests from your application or system | Best for custom integrations |
| MCP | Tool calls from a compatible AI client | Best for conversational access to supported operations |
GeeLark CLI provides command-line access to supported GeeLark operations using underlying API capabilities. Developers, scripts, and compatible AI agents can run CLI commands directly. The same geelark-cli package can also run as an MCP server, allowing MCP-compatible AI clients to access supported operations through structured tool calls. Some browser profile operations require a separate local Browser API connection.
Run your first GeeLark command
Check your GeeLark API access and follow the installation requirements for your chosen environment.
Install the CLI
For the npm installation method, use a working Node.js and npm environment.
$ npm install -g geelark-cli@latestConfigure authentication locally
Start the interactive configuration and provide your GeeLark API token on your computer.
$ geelark-cli config initRetrieve your cloud phone list
Run a simple query and compare the returned information with your GeeLark workspace.
$ geelark-cli phone list --page 1 --page-size 10Find connection information in GeeHub. Keep your token private. Browser operations may require a configured local Browser API connection.
CLI documentation and related connections
Reference material for the CLI and its sibling connection methods.
Official GeeLark CLI documentation
Installation, configuration, and the full command reference for geelark-cli.
GeeLark API for system integrations
Send requests from your application or system to supported GeeLark operations.
Connect AI tools with GeeLark MCP
Expose supported operations as tools for compatible AI clients.
Frequently Asked Questions
What is GeeLark CLI?
GeeLark CLI is the official command-line tool for supported GeeLark operations. You can run commands directly from a terminal or include them in your scripts.
Can I manage cloud phones and browser profiles?
Yes. The CLI includes commands for both. Cloud phone and browser operations may use different connections and have different runtime requirements.
Can I use CLI commands in scripts?
Yes. You can incorporate supported commands into your scripts and process their output. Your script is responsible for any additional logic, checks, and error handling.
Which output formats are available?
The official CLI supports JSON, pretty-printed JSON, and table output through the –format option.
Can I manage automation tasks?
The CLI provides documented commands for supported task operations, including task queries, history, and available custom-task workflows. Consult the command reference for the exact options.
How is CLI different from API and MCP?
CLI provides terminal commands. API enables direct integration through programmatic requests. MCP exposes supported operations as tools for compatible AI clients.
Can AI agents use GeeLark CLI?
Compatible AI agents that can execute terminal commands may use the CLI. The package also supports MCP server mode for MCP-compatible clients.
What do I need to get started?
You need GeeLark API access, a locally configured token, and an installation method supported by your environment. Check the official documentation for operating-system support and browser connection requirements.
Bring GeeLark into your terminal workflow
Run supported cloud phone, browser profile, and automation operations from the tools you already use.
Image to video

