Firecrawl Skills plugin
Firecrawl skills catalog for Claude Code, Cursor and Codex: web search, scraping, crawling, page interaction and API integration
- 155 ★ on the repository
- Last commit
- ISC
Extended access
Why this level
- Skills reach the web through the Firecrawl service and open pages from links in the task
- The FIRECRAWL_API_KEY key is kept in the project environment
- The interact skill clicks and fills forms on websites
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/firecrawl-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Skills reach the web through the Firecrawl service and open pages from links in the task
- The FIRECRAWL_API_KEY key is kept in the project environment
- The interact skill clicks and fills forms on websites
This is third-party code. Review the repository files before installing.
What it does
Firecrawl Skills is a plugin with the full set of skills for Firecrawl, a service that pulls web data for AI agents. It ships in three forms: for Claude Code, for Cursor and for Codex. There are three groups. The core group teaches an agent to use the Firecrawl CLI during a live session: search the web, fetch a page, crawl a site, map its URLs, click and fill forms on a loaded page, monitor changes, parse files and download them. Next to them sit indexes for finding research papers (PubMed, bioRxiv, medRxiv, arXiv) and for answering developer questions from issues, pull requests and documentation. The build group explains how to wire the Firecrawl API into product code: pick an SDK or REST, choose an endpoint, put the key into the project. The workflows group holds ready recipes: lead generation, deep research, SEO audit, competitive intelligence, job postings, company directories. The plugin also configures the Firecrawl MCP server, so editors with MCP support get the same tools. The catalog is a mirror synced from Firecrawl source repositories; skill changes are made there.
Who it is for. Developers and analysts who hand web data collection to an agent and build Firecrawl into their own products.
Good fit when
- The agent has to find and read current web pages during a session
- You need to crawl a whole documentation site and build a knowledge base from it
- You are adding scraping or search to a project through the Firecrawl API and need an SDK choice
- You want a ready recipe such as an SEO audit or competitor research
Not a fit when
- The data sits behind sources you have no right to read
- One ordinary search query solves the task without crawling pages
- You want to change skills: edits are accepted in the source repositories, not here
- You have no Firecrawl key and no self-hosted instance of the service
Example request
Use Firecrawl to find fresh articles about the new Next.js release, read the best three and write a short summary with linksLimitations
A Firecrawl account or your own instance is required, and the key lives in the FIRECRAWL_API_KEY variable. Requests to sites go through the service, and volume and price follow the Firecrawl plan. Skills in this repository are synced automatically, so manual edits are overwritten at the next sync. Treat third-party content that skills fetch from the web as untrusted. Working through the service means the contents of fetched pages and your requests pass through Firecrawl, and limits depend on the account plan.
How to disable. Remove the plugin with /plugin in Claude Code or the skills with npx skills remove. Delete the FIRECRAWL_API_KEY key from the project environment.
FAQ
How do I install the Firecrawl skills in Claude Code?
Run npx skills add firecrawl/skills: without flags it opens an interactive picker, and --skill installs one skill by name, for example firecrawl-build. A Claude Code plugin also lives in the .claude-plugin folder.
How does core differ from build?
Core teaches the agent to use Firecrawl inside the session through the CLI or MCP: search, fetch pages, crawl sites. Build helps write product code that calls the Firecrawl API itself and guides the choice of SDK and endpoints.
Where do skill changes go?
The skills folders here are mirrored automatically. Core skills are edited in firecrawl/cli, build skills in firecrawl/firecrawl and workflows in firecrawl/firecrawl-workflows.
Do I need an API key?
Yes. For cloud Firecrawl the key goes into the FIRECRAWL_API_KEY variable and is issued at firecrawl.dev/app. A self-hosted instance of the service also works.
What does the editor plugin include?
The plugin bundles the core and build skills for Claude Code, Cursor and Codex and wires in the Firecrawl MCP server. Workflow recipes are not bundled: the author treats them as session recipes.
Discussion
No questions or reviews yet.
Related
PluginsThe official Firecrawl plugin for finding sources, reading web pages and researching companies in Claude Code
The Firecrawl command-line client with agent skills: web search, scraping, crawling, URL mapping and page interaction
Five skills that teach an agent to wire Firecrawl search, scraping and page interaction into the code of your own application
Official Firecrawl MCP server: web search, page scraping to Markdown, site crawling and in-page actions
Ready-made Firecrawl workflows: sourced research, SEO audits, site QA, knowledge bases and company contact lists
Portable skill for the Crawl4AI CLI and SDK: JavaScript-heavy pages, structured extraction by CSS schemas and batch crawling.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Firecrawl Skills (this card) | yes | 155 | ||
| Firecrawl Build Skills | no | 201,311 | only the skills for wiring the API into product code | |
| Firecrawl Workflows | no | 224 | only the ready end-to-end recipes | |
| Firecrawl MCP Server | yes | API key | 7,809 | the MCP server alone, without skills |
README in short
The README calls the repository a distribution catalog: a table of three categories, npx skills install commands, the core and build skill lists, workflow recipes and MCP wiring. It also lists supported editors and the key requirement.