Manage Ranla campaigns, segments, and other workspace tasks from the terminal while keeping sign-in credentials private and consequential actions under user approval.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ranla-cli
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install ranla-cli using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Ranla CLI?
Ranla CLI lets an AI agent work with a Ranla account through Ranla's own command interface, including campaigns, segments, pages, ads, and Reddit workflows. Instead of asking for an API key, the agent starts a browser-based sign-in flow; the CLI stores its credential locally, so the agent never sees it.
Designed for safe, user-directed operations, the skill encourages reading command schemas and drafting or proposing changes before writing. Any action that sends, publishes, or spends must be confirmed by the user. As one of the Openclaw Skills for marketing operations, it combines terminal-based access with explicit workspace selection and approval controls.
Ranla CLI Use Cases
- List or draft Ranla campaigns and inspect their available fields before making changes.
- Build or review audience segments using Ranla's native commands.
- Work with Ranla pages, ads, Reddit tasks, and other supported command areas.
- Check pending approval cards and present their summaries and app links to the user.
- Sign in to a selected Ranla workspace without requesting or exposing an API key.
- Inspect which model-key configuration Ranla uses through the
braincommand.
How Ranla CLI Works
- Start browser-based authentication with
npx -y ranla-cli@latest login; the CLI returns a verification URL and user code. - Give the user the URL and code. The user signs in, confirms the code, selects a workspace, and approves access.
- After the user confirms approval, run
loginagain and verify the signed-in identity and workspace withwhoami. - If the account has multiple workspaces, list them and ask the user which one to use; select it explicitly or pass
--workspacefor a single call. - Inspect the available commands and their input schemas with
run helpbefore invoking a command. Submit one JSON object through--input,--file, or standard input. - Review the JSON result. For actions involving sending, publishing, or spending, show the approval summary and ask the user before approving or rejecting the pending card.
- Log out when requested to revoke the CLI credential and remove it from the computer.
Ranla CLI Setup
Use the latest package version for each invocation, or use ranla if the user has installed the CLI globally.
npx -y ranla-cli@latest login
Share the returned verification_url and user_code with the user. Do not request their password or API key. After the user approves in the browser, finish sign-in and check the active account:
npx -y ranla-cli@latest login
npx -y ranla-cli@latest whoami
For accounts with multiple workspaces, let the user choose before selecting one:
npx -y ranla-cli@latest workspaces list
npx -y ranla-cli@latest workspaces use "<name>"
Discover commands and required fields before operating, then provide input as a JSON object:
npx -y ranla-cli@latest run help
npx -y ranla-cli@latest run help <noun>
npx -y ranla-cli@latest run <noun> <verb> --input '{"field":"value"}'
Input can also come from a file or standard input using --file input.json or --input -. When output is piped, results are JSON. Exit code 0 indicates success, 1 indicates an operation failure, and 2 indicates invalid command-line syntax; use the error message to correct the command rather than guessing. To sign out:
npx -y ranla-cli@latest logout
Ranla CLI Data Schema & Taxonomy
The skill does not define a separate local project database or generated file format. It uses Ranla CLI commands and JSON input/output.
| Data or metadata | Organization and handling |
|---|---|
| Command inputs | A single JSON object passed with --input, --file, or stdin. Use run help <noun> to inspect the exact schema. |
| Command results | JSON when piped; failures provide a message on stderr. |
| Sign-in state | The CLI saves its credential locally after browser approval. The agent must not print, log, or commit credentials. |
| Workspace context | Each sign-in corresponds to a workspace. Users can list and select workspaces, or specify one for a single call with --workspace "<name>". |
| Approval cards | Pending actions include a message ID, summary, and app URL; inspect them with approvals list or approvals get <id>. |
| Model configuration | brain reports whether Ranla uses Included or a workspace-provided OpenRouter, Anthropic, OpenAI, or Google key. Adding a key is the user's responsibility. |
Ranla CLI Advanced Features
- Browser-based sign-in avoids asking the user to paste an API key; credentials remain managed by the CLI.
- Workspace-aware operation supports explicit selection and per-call workspace targeting. The agent must not choose a workspace on the user's behalf.
- Command discovery exposes available nouns, verbs, and input fields, helping prevent guessed or malformed operations.
- Approval workflows let the agent list or inspect pending cards and approve or reject them only after the user decides. Some payout approvals must be completed in the app.
- Safety guidance requires user confirmation before actions that cost money, send messages, or publish content, even if a command does not pause for approval.
- The
braincommand reports Ranla's model-key setup. Users must add provider keys themselves; the agent must never request, expose, or pass keys through commands. logoutrevokes the CLI credential and removes it from the current computer.
SKILL.md
Loading
Related Openclaw Skills
Qinghu Video Channel Comments
Fetch comments for a WeChat Video Channel work, paginate through results, and expand replies to a selected comment.
Qinghu Video Channel Live Room Details
Retrieve details for a WeChat Channels live room by live ID through Qinghu AI’s qhkit tool.
QinghuAI Video Channel Collections
Look up the collections published by a WeChat Channels account and return the topic IDs and types needed for follow-up video queries.
Qinghu Video Channel Collection Videos
Retrieve all videos in a Qinghu Video Channel collection by topic ID with cursor-based pagination and structured results.
Remember Me Bible Memory
Remember Me Bible Memory connects an AI assistant to spaced-repetition tools for adding, reviewing, reading, and organizing Bible verses.
Whiteboard
Whiteboard lets you and an AI coding agent create, inspect, and iteratively refine persistent local Excalidraw diagrams.
Featured*
