# Extract Web Data with Firecrawl

Teams often need web content in clean formats without writing custom scrapers. This skill guides Claude, Codex, and Claude Code through Firecrawl modes for scraping, search, maps, crawling, extraction, and status checks.

## Install

```bash
npx skillstore add ck991357/firecrawl
```

## Metadata

- Status: approved
- Slug: ck991357-firecrawl
- Version: 1
- Author version: 1
- Skillstore revision: r1
- Version status: invalid
- Tree hash: d516fc58fa69830298b68f0f4371571c0b1c500016fc8cacdd8463a4d35d327d
- Author: CK991357
- GitHub username: CK991357
- License: MIT
- Repository: https://github.com/CK991357/gemini-chat/tree/main/src/skills/firecrawl
- Ref: 02f077c174c5335e2f5d02ca15e77b70d9543e58
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network
- Quality score: 78
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/ck991357-firecrawl
- Manifest: https://skillstore.pages.dev/api/skills/ck991357-firecrawl/manifest

## Capabilities

- Shows how to scrape a single web page and request markdown or HTML output.
- Shows how to run web searches with a query and result limit.
- Shows how to request a site map from a target URL.
- Shows how to start asynchronous crawl jobs with a crawl limit.
- Shows how to extract structured fields from one or more URLs.
- Shows how to check an asynchronous job with a job identifier.

## Use Cases

- Collect Research Sources: Gather page content, search results, and site maps for topic research without writing scraper code.
- Prepare Content for Analysis: Convert web pages into markdown or HTML before summarization, comparison, or extraction tasks.
- Extract Fields from Articles: Use extraction prompts and schemas to capture titles, authors, dates, and other structured fields.

## Prompt Templates

### Scrape One Page

```
Use Firecrawl scrape mode for this URL. Return the page content in markdown and note any access problems.
```

### Search a Topic

```
Use Firecrawl search mode for this topic with a limit of five results. Summarize the most relevant pages.
```

### Map a Website

```
Use Firecrawl map mode for this website. Group discovered URLs by likely purpose and flag pages worth scraping.
```

### Run Structured Extraction

```
Use Firecrawl extract mode on these URLs. Extract title, author, publish date, summary, and source URL, then check job status.
```

## Limitations

- Requires the Firecrawl tool to be installed and available in the agent environment.
- Documents tool calls only; it does not include a Firecrawl implementation.
- Asynchronous crawl and extract modes require a separate status check.
- Results depend on target site access, robots rules, and Firecrawl service limits.

## Best Practices

- Use full URLs with http or https, because Firecrawl expects complete URL formats.
- Keep crawl limits small until you understand the target site size and result quality.
- Use status checks for crawl and extract jobs before assuming asynchronous work is complete.

## Anti Patterns

- Do not place mode-specific parameters at the top level of the request.
- Do not pass a single string when extract mode expects an array of URLs.
- Do not crawl large sites without clear limits, scope, and permission to access the content.

## Security Audit

- Audited at: 2026-07-05T06:36:03.625\+00:00
- Summary: The external command findings are false positives from Markdown backticks and fenced JSON examples, not executable code. The network findings are example URLs and URL-format guidance for Firecrawl usage. No prompt injection, credential access, or covert exfiltration instructions were found.

## Stats

- Views: 271
- Downloads: 18
- Favorites: 1
- Popularity score: 0