PhantomBuster Skill
name: phantombuster
by capt-marbles · published 2026-03-22
$ claw add gh:capt-marbles/capt-marbles-phantombuster---
name: phantombuster
description: Control PhantomBuster automation agents via API. List agents, launch automations, get output/results, check status, and abort running agents. Use when the user needs to run LinkedIn scraping, Twitter automation, lead generation phantoms, or any PhantomBuster workflow.
version: 1.0.0
author: captmarbles
---
# PhantomBuster Skill
Control your [PhantomBuster](https://phantombuster.com) automation agents from the command line.
Setup
1. Get your API key from [Workspace Settings](https://phantombuster.com/workspace-settings)
2. Set the environment variable:
```bash
export PHANTOMBUSTER_API_KEY=your-api-key-here
```
Usage
All commands use the bundled `pb.py` script in this skill's directory.
List Agents
See all your configured PhantomBuster agents.
python3 pb.py list
python3 pb.py list --json # JSON outputLaunch an Agent
Start a phantom by ID or name.
python3 pb.py launch <agent-id>
python3 pb.py launch <agent-id> --argument '{"search": "CEO fintech"}'Get Agent Output
Fetch the results/output from the most recent run.
python3 pb.py output <agent-id>
python3 pb.py output <agent-id> --json # Raw JSONCheck Agent Status
See if an agent is running, finished, or errored.
python3 pb.py status <agent-id>Abort Running Agent
Stop an agent that's currently running.
python3 pb.py abort <agent-id>Get Agent Details
Full details about a specific agent.
python3 pb.py get <agent-id>Fetch Result Data
Download the actual result data (CSV) from an agent's latest run.
python3 pb.py fetch-result <agent-id>
python3 pb.py fetch-result <agent-id> > output.csvThis downloads the `result.csv` file from the agent's S3 storage, perfect for integrating PhantomBuster data into your workflows.
Example Prompts
Common Phantoms
PhantomBuster offers many pre-built automations:
Rate Limits
PhantomBuster has execution time limits based on your plan. The API itself is not heavily rate-limited, but agent execution consumes your plan's minutes.
More tools from the same signal band
Order food/drinks (点餐) on an Android device paired as an OpenClaw node. Uses in-app menu and cart; add goods, view cart, submit order (demo, no real payment).
Sign plugins, rotate agent credentials without losing identity, and publicly attest to plugin behavior with verifiable claims and authenticated transfers.
The philosophical layer for AI agents. Maps behavior to Spinoza's 48 affects, calculates persistence scores, and generates geometric self-reports. Give your...