Firecrawl Tool icon

Firecrawl Tool

Makes a web scraping request using Firecrawl and returns the scraped data

Scrape Url

AI-generated

Summary

Scrape content from a specified URL using the Firecrawl service, allowing customization of output formats and content filtering, with optional caching to speed up repeated requests.

Inputs

  • toolDescription — Text description to explain to the AI what this tool does. Defaults to 'Scrapes content from a given URL using Firecrawl'.
  • url (required) — The URL to scrape content from. Can include placeholders that are dynamically replaced during execution.
  • formats — The formats for the scraped output. Options include Markdown, HTML, Raw HTML, Content, Links, Screenshot, Full Page Screenshot, Extract, and JSON. Defaults to Markdown.
  • onlyMainContent — Boolean indicating whether to return only the main content excluding headers, navigation bars, footers, etc. Defaults to true.
  • includeTags — Comma-separated list of tags, classes, or IDs to explicitly include in the output.
  • excludeTags — Comma-separated list of tags, classes, or IDs to exclude from the output.
  • cache — Caching method to speed up scraping. Supports 'none' or 'postgres'. Defaults to 'postgres'.
  • cacheTTL — Time to live for the cache in seconds when using Postgres caching. Defaults to -1 (no expiration). Only shown if 'postgres' caching is enabled.
  • placeholderDefinitions — Definitions for placeholders used in the URL or tags, specifying name, description, and type (string or URL). Used to validate and replace dynamic placeholders.

Output shape

a single AI tool instance that performs the scraping function and returns scraped content formatted as specified

The output is a JSON-stringified response containing the scraped data in selected formats. Supports caching of scrape results when Postgres caching is enabled with an optional TTL. The operation requires a Firecrawl API credential and, if using Postgres caching, a Postgres credential.

Links

Discussion