A comprehensive web scraping and automation toolkit that enables AI agents to crawl websites, map URLs, and execute browser actions via a cloud-based sandbox.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install firecrawl-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 firecrawl-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 Firecrawl CLI?
The Firecrawl CLI skill is a powerful integration designed to bridge the gap between AI agents and the live web. It provides a robust set of tools for scraping single URLs into clean markdown, crawling entire websites asynchronously, and performing deep web searches. By using this tool within the ecosystem of Openclaw Skills, developers can empower their agents to extract high-quality data while bypassing complex anti-bot mechanisms and rendering challenges.
Beyond simple data extraction, this skill offers a sophisticated cloud browser sandbox. This allows agents to perform complex automation tasks such as clicking elements, taking snapshots, and executing natural language queries to navigate websites like a human user. Whether you are building a knowledge base or an automated research assistant, this skill provides the necessary infrastructure for reliable web interactions.
Firecrawl CLI Use Cases
- Extracting clean markdown from documentation pages to feed into LLM context windows.
- Mapping the entire link structure of a domain to identify relevant sub-pages for deep analysis.
- Running automated web searches that return scraped content instead of just a list of links.
- Orchestrating multi-step browser interactions, such as logging into a portal or navigating complex SPAs.
- Building automated data pipelines that crawl sites periodically to monitor for changes.
How Firecrawl CLI Works
- The user installs the firecrawl-cli and authenticates via an API key or browser-based login.
- Commands are issued to either scrape a specific URL or initiate a long-running crawl job.
- The Firecrawl engine handles JavaScript rendering and content extraction, filtering out noise like headers and footers.
- For browser automation, a cloud session is launched, providing a persistent sandbox for executing sequential actions.
- The processed data is returned in structured formats like JSON or Markdown, ready for integration into the Openclaw Skills workflow.
Firecrawl CLI Setup
To get started, install the CLI globally via npm:
npm install -g firecrawl-cli
Authenticate your environment using the browser login (recommended for agents) or by setting an environment variable:
firecrawl login --browser
# or
export FIRECRAWL_API_KEY=fc-YOUR-KEY
Verify your setup and check your available concurrency credits by running:
firecrawl --status
Firecrawl CLI Data Schema & Taxonomy
The skill manages data across several output formats to ensure compatibility with different agent workflows. Below is a summary of the data organization:
| Feature | Output Type | Description |
|---|---|---|
| Scrape | Markdown/JSON | The primary content of the page, optionally including links and metadata. |
| Crawl | JSON Object | A job-based structure containing a list of all successfully scraped URLs and their content. |
| Map | String Array | A simple list of discovered URLs found during a site discovery process. |
| Browser | @ref IDs | Unique element identifiers used to reference DOM elements during automation sessions. |
| Agent | Text/JSON | Natural language responses derived from web searching and browsing tasks. |
Firecrawl CLI Advanced Features
- Smart content extraction using the --only-main-content flag to eliminate navigation and footer clutter.
- Asynchronous crawling with job ID tracking, allowing agents to poll for results on large-scale tasks.
- Multi-format scraping support, enabling the simultaneous retrieval of markdown, HTML, and metadata.
- Integration with AI agents for natural-language driven web navigation and data gathering.
- Support for self-hosted Firecrawl instances by configuring the --api-url parameter.
SKILL.md
Loading
Related Openclaw Skills
Gemini Web Ops
An advanced automation skill that interacts with the Gemini web interface to perform high-level reasoning and AI image generation.
Lecture Notes Master
A powerful AI skill that converts various media into deep, multi-layered Obsidian vaults using recursive atomic decomposition.
Skill Inventory Manager
An automated utility to scan, index, and manage your library of installed Openclaw Skills for AI agents.
x-mirror
A comprehensive CLI tool to manage and switch download mirrors for various system and language package managers.
Commodities Quote
A high-precision financial skill for retrieving real-time commodity prices and technical market indicators via Octagon MCP.
Turing Tavily Web Search
A powerful web search skill for AI agents that leverages the Turing Tavily proxy to fetch real-time information and extracted web content.
Featured*
