Tines

Tines docs: HTTP Request action and CredentialsAdd live web data from Firecrawl to Tines stories

Overview

Tines stories can call any REST API with the HTTP Request action. Pointing that action at the Firecrawl API lets a story search the web, scrape a page into markdown, pull structured data, map a site, or crawl it, and pass the result to the next action in the story.

Quick start

Create a Firecrawl API key
In Tines, open Credentials, create a Text credential named firecrawl, and paste the key as its value
Add an HTTP Request action to your story and configure it as shown below

Referencing the key as <<CREDENTIAL.firecrawl>> keeps it out of the action configuration.

Scrape a page

This action sends a scrape request and emits the page as markdown:

{
  "url": "https://api.firecrawl.dev/v2/scrape",
  "method": "post",
  "content_type": "json",
  "headers": {
    "Authorization": "Bearer <<CREDENTIAL.firecrawl>>"
  },
  "payload": {
    "url": "https://example.com",
    "formats": ["markdown"],
    "integration": "_tines"
  }
}

Replace https://example.com with a URL from an upstream event, for example <<receive_event.body.url>>. The optional integration field labels requests that come from Tines.

What you can do

Capability	Endpoint	Use case
Search	`POST /v2/search`	Find sources for a question or alert before later actions
Scrape	`POST /v2/scrape`	Turn a URL into markdown or structured JSON
Extract	`POST /v2/scrape` with a `json` format	Pull fields such as a title or product details with a prompt and schema
Map	`POST /v2/map`	List the URLs on a site
Crawl	`POST /v2/crawl`	Gather pages across a site, then poll for the result

All endpoints use the base URL https://api.firecrawl.dev, the same Authorization header, and a JSON body.

Search the web

{
  "url": "https://api.firecrawl.dev/v2/search",
  "method": "post",
  "content_type": "json",
  "headers": {
    "Authorization": "Bearer <<CREDENTIAL.firecrawl>>"
  },
  "payload": {
    "query": "firecrawl web scraping",
    "limit": 5,
    "sources": ["web"],
    "integration": "_tines"
  }
}

Extract structured data

Set the formats entry to the json type with a prompt and a JSON schema:

{
  "url": "https://api.firecrawl.dev/v2/scrape",
  "method": "post",
  "content_type": "json",
  "headers": {
    "Authorization": "Bearer <<CREDENTIAL.firecrawl>>"
  },
  "payload": {
    "url": "https://example.com",
    "formats": [
      {
        "type": "json",
        "prompt": "Extract the page title",
        "schema": {
          "type": "object",
          "properties": { "title": { "type": "string" } }
        }
      }
    ],
    "integration": "_tines"
  }
}

Crawl a site

A crawl runs asynchronously. Use one action to start it and a second action to read the result. Start the crawl with POST /v2/crawl:

{
  "url": "https://api.firecrawl.dev/v2/crawl",
  "method": "post",
  "content_type": "json",
  "headers": {
    "Authorization": "Bearer <<CREDENTIAL.firecrawl>>"
  },
  "payload": {
    "url": "https://example.com",
    "limit": 10,
    "integration": "_tines"
  }
}

The response includes an id. Check the status with a second HTTP Request action that reads the crawl by that id, and repeat it until the status is completed:

{
  "url": "https://api.firecrawl.dev/v2/crawl/<<start_crawl.body.id>>",
  "method": "get",
  "headers": {
    "Authorization": "Bearer <<CREDENTIAL.firecrawl>>"
  }
}

Here start_crawl is the name of the action that started the crawl.

Get Started

Core Endpoints

More

Quickstarts

Developer Guides

Webhooks

Use Cases

Other

Contributing

Overview

Quick start

Scrape a page

What you can do

Search the web

Extract structured data

Crawl a site

Resources

​Overview

​Quick start

​Scrape a page

​What you can do

​Search the web

​Extract structured data

​Crawl a site

​Resources

Overview

Quick start

Scrape a page

What you can do

Search the web

Extract structured data

Crawl a site

Resources