> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mantle.chat/llms.txt
> Use this file to discover all available pages before exploring further.

# Firecrawl

> Scrape and crawl websites to extract structured content — power your AI agents with real-time web data from Mantle

Connect your Firecrawl API key and let your AI agents scrape web pages, crawl entire sites, extract structured data, and
perform deep research — turning the web into a data source for your workflows.

***

## What you unlock

<CardGroup cols={2}>
  <Card title="Web scraping" icon="globe">
    Scrape any URL and extract clean, structured content — ready for AI processing and analysis.
  </Card>

  <Card title="Site crawling" icon="spider">
    Crawl entire websites recursively to discover pages, map site structure, and extract content at scale.
  </Card>

  <Card title="Deep research" icon="microscope">
    Run multi-step research tasks that follow links, gather data, and synthesize findings across the web.
  </Card>

  <Card title="Structured extraction" icon="table">
    Extract specific data fields from web pages using schemas — product prices, article metadata, contact info, and
    more.
  </Card>
</CardGroup>

***

## Connect Firecrawl

<Steps>
  <Step title="Open Settings → Integrations">
    In any workspace go to **Settings → Integrations** and select **Firecrawl**.
  </Step>

  <Step title="Get your API key">
    Sign up or log in at [firecrawl.dev](https://firecrawl.dev), navigate to your dashboard, and copy your API key.
  </Step>

  <Step title="Paste and connect">Paste the key into the **API Key** field and click **Connect**.</Step>
</Steps>

***

## Agent tools

Add Firecrawl tools to agents that need them. Their access is limited by the connected key and selected tools.

### Scraping

| Tool         | Description                                       |
| ------------ | ------------------------------------------------- |
| Scrape       | Scrape a single URL and extract its content       |
| Batch scrape | Scrape multiple URLs in a single request          |
| Extract      | Extract structured data from a URL using a schema |

### Crawling & Mapping

| Tool             | Description                                     |
| ---------------- | ----------------------------------------------- |
| Crawl            | Crawl a website recursively starting from a URL |
| Get crawl status | Check the status of a running crawl job         |
| Map URLs         | Map the URL structure of a website              |
| Search           | Search the web and return structured results    |

### Deep Research & AI Agents

| Tool                | Description                                     |
| ------------------- | ----------------------------------------------- |
| Deep research       | Run a multi-step research task across the web   |
| Get research status | Check the status of a deep research task        |
| Start agent         | Launch an AI agent to perform complex web tasks |
| Get agent status    | Check the status of a running agent task        |

### LLMs.txt & Usage

| Tool              | Description                             |
| ----------------- | --------------------------------------- |
| Generate LLMs.txt | Generate an LLMs.txt file for a website |
| Get credit usage  | Check your Firecrawl credit consumption |

***

## Triggers

<Note>
  Automated triggers for Firecrawl are not yet available. Triggers will be added in a future update — check back soon.
</Note>

***

## Example use cases

<AccordionGroup>
  <Accordion title="Competitor research on demand">
    Ask your agent to scrape a competitor's pricing page and extract their plan names, prices, and feature lists into a
    structured table — instant competitive intelligence without manual research.
  </Accordion>

  <Accordion title="Content aggregation for knowledge bases">
    Crawl your documentation site and extract all pages into structured content. Use the results to build a searchable
    knowledge base or feed it into another AI system.
  </Accordion>

  <Accordion title="Deep research on a topic">
    Ask your agent to research a topic like "serverless database trends" — Firecrawl's deep research follows links
    across multiple sources and synthesizes a comprehensive report.
  </Accordion>
</AccordionGroup>

***

## Troubleshooting

<AccordionGroup>
  <Accordion title="My API key isn't working">
    Verify the key was copied from [firecrawl.dev](https://firecrawl.dev) and hasn't expired. Generate a fresh key if
    the current one doesn't work.
  </Accordion>

  <Accordion title="Scrape returns empty content">
    Some websites use heavy JavaScript rendering that may not be captured by default. Check Firecrawl's documentation
    for options like enabling JavaScript rendering or wait-for selectors.
  </Accordion>

  <Accordion title="Crawl job seems stuck">
    Large sites can take time to crawl. Use "Get crawl status" to check progress. If the job is truly stuck, cancel it
    and retry with a smaller scope (e.g., limit max pages or depth).
  </Accordion>

  <Accordion title="Credit usage is higher than expected">
    Each scrape, crawl page, and research step consumes credits. Use "Get credit usage" to monitor consumption and
    consider batching operations to reduce overhead.
  </Accordion>
</AccordionGroup>

***

<Card title="Back to Integrations" icon="plug" href="/integrations/overview">
  View all available integrations
</Card>