With these Public Integrations, you can search web pages, retrieve result snippets, and extract content in multiple formats such as markdown, HTML, links, screenshots, or PDF files.
-
com.globant.geai.firecrawl.web_scraper: Scrapes content from a given web page and can handle many sites that attempt to block scraping. It can also capture full-page screenshots and retrieve and parse PDFs available on the web. You must provide a URL or you can use it together with com.globant.geai.firecrawl.web_search.
-
com.globant.geai.firecrawl.web_search: Searches the web and returns short text snippets from the top results. It is recommended to use it together with com.globant.geai.firecrawl.web_scraper.
Before using these Tools, you must obtain an API key from Firecrawl and register it as a Credential in your Project.
- Go to https://www.firecrawl.dev/app and sign in. Your API key is created automatically and shown in your dashboard. Copy the key.
- In the Integrations Dashboard, open the Credentials tab and click on + New Credential.
- In the Name field, enter a descriptive name (for example, firecrawl_api_token).
- In the Integration dropdown, select Firecrawl Web.
- Note that the Authentication Type is automatically set to API Key, and the Authentication Level is automatically set to Project. These values are predefined and cannot be modified.
- In API Key Value, paste the Firecrawl API key exactly as provided.
- Confirm to save the Credential.
The Credential is now associated with the Firecrawl Integration at the Project level and will be used by both com.globant.geai.firecrawl.web_scraper and com.globant.geai.firecrawl.web_search.
After creating the Credential, you can add either Tool to your Agent from the AI & Tools Tab. No additional authentication parameters need to be configured in the Agent, as the Integration automatically resolves the active Credential at runtime.