# Firecrawl MCP

> Scrape, crawl, map, search, and extract web data with the Firecrawl API.

Lets an agent turn the web into clean data - scrape a single page, crawl a whole
site, map a site to a URL list, search the web with full-page content, and extract
structured data with natural language.

Product page: [flymy.ai/mcp/firecrawl](https://flymy.ai/mcp/firecrawl/)

## What it can do

| Method | What it does |
| --- | --- |
| `firecrawl_scrape` | Scrape one page and return markdown, HTML, links, a summary, a screenshot, or schema-guided JSON. |
| `firecrawl_crawl` | Start crawling a whole site (async job); returns an id to poll. |
| `firecrawl_get_crawl_status` | Check a crawl job and fetch the pages scraped so far. |
| `firecrawl_cancel_crawl` | Cancel a running crawl job. |
| `firecrawl_crawl_params_preview` | Preview the crawl parameters Firecrawl derives from a prompt, without crawling. |
| `firecrawl_map` | Map a site into a list of URLs quickly, with optional search filter. |
| `firecrawl_search` | Search the web and optionally scrape each result's full content. |
| `firecrawl_extract` | Start extracting structured data from pages by prompt and/or schema (async job). |
| `firecrawl_get_extract_status` | Check an extract job and fetch the structured result. |
| `firecrawl_batch_scrape` | Start a batch scrape of many URLs in one async job. |
| `firecrawl_get_batch_scrape_status` | Check a batch scrape job and fetch results so far. |
| `firecrawl_get_credit_usage` | Get the team's remaining credits. |
| `firecrawl_get_token_usage` | Get the team's remaining tokens and billing period. |

## How to get credentials

None for `firecrawl_scrape`, `firecrawl_map` and `firecrawl_search`: on FlyMyAI
Cloud they run on FlyMyAI's Firecrawl key and each call is billed to your
FlyMyAI balance.

Crawls, batch scrapes, extract jobs and the credit-usage reads need your own
Firecrawl key (they take credits page by page after the call that starts them,
and the usage endpoints read the account the key belongs to):

1. Open [firecrawl.dev/app/api-keys](https://www.firecrawl.dev/app/api-keys).
2. Sign in (or create a free Firecrawl account).
3. Create a new API key and copy the `fc-...` value.
4. Paste it into the Firecrawl connection in FlyMyAI as `FIRECRAWL_API_KEY`.

With your own key Firecrawl bills your account; `firecrawl_get_credit_usage`
shows what is left.

## Fields to fill in FlyMyAI

| FlyMyAI field | Where it comes from |
| --- | --- |
| `FIRECRAWL_API_KEY` (optional; needed for crawl, batch scrape, extract and usage) | Dashboard -> API Keys -> the `fc-...` key |

## Troubleshooting

- **`401 Unauthorized`** - the key is missing or wrong. Recopy the `fc-...` key.
- **`402 Payment Required`** - out of credits. Check `firecrawl_get_credit_usage`
  or upgrade the plan.
- **Crawl / extract still pending** - these are async jobs. Poll
  `firecrawl_get_crawl_status` / `firecrawl_get_extract_status` with the returned id.

## Links

- [Firecrawl API keys](https://www.firecrawl.dev/app/api-keys)
- [API reference](https://docs.firecrawl.dev/api-reference/v2-introduction)