Skip to main content
Firecrawl /agent is a magic API that searches, navigates, and gathers data from even the most complex websites, finding data in hard-to-reach places and discovering data anywhere on the internet. It accomplishes in a few minutes what would take a human many hours, and makes traditional web scraping obsolete. Just describe what data you want and /agent handles the rest.
Research Preview: Agent is in early access. Expect rough edges. It will get significantly better over time. Share feedback →
Agent builds on everything great about /extract and takes it further:
  • No URLs Required: Just describe what you need via prompt parameter. URLs are optional.
  • Deep Web Search: Autonomously searches and navigates deep into sites to find your data
  • Reliable and Accurate: Works with a wide variety of queries and use cases
  • Faster: Processes multiple sources in parallel for quicker results
  • Cheaper: Agent is more cost-effective than /extract for complex use cases

​
Using /agent

The only required parameter is prompt. Simply describe what data you want to extract. For structured output, provide a JSON schema. The SDKs support Pydantic (Python) and Zod (Node) for type-safe schema definitions: 
from firecrawl import FirecrawlApp
from pydantic import BaseModel, Field
from typing import List, Optional

app = FirecrawlApp(api_key="fc-YOUR_API_KEY")

class Founder(BaseModel):
    name: str = Field(description="Full name of the founder")
    role: Optional[str] = Field(None, description="Role or position")
    background: Optional[str] = Field(None, description="Professional background")

class FoundersSchema(BaseModel):
    founders: List[Founder] = Field(description="List of founders")

result = app.agent(
    prompt="Find the founders of Firecrawl",
    schema=FoundersSchema
)

print(result.data)

​
Response

JSON
{
  "success": true,
  "status": "completed",
  "data": {
    "founders": [
      {
        "name": "Eric Ciarla",
        "role": "Co-founder",
        "background": "Previously at Mendable"
      },
      {
        "name": "Nicolas Camara",
        "role": "Co-founder",
        "background": "Previously at Mendable"
      },
      {
        "name": "Caleb Peffer",
        "role": "Co-founder",
        "background": "Previously at Mendable"
      }
    ]
  },
  "expiresAt": "2024-12-15T00:00:00.000Z",
  "creditsUsed": 15
}

​
Providing URLs (Optional)

You can optionally provide URLs to focus the agent on specific pages: 
from firecrawl import FirecrawlApp

app = FirecrawlApp(api_key="fc-YOUR_API_KEY")

result = app.agent(
    urls=["https://docs.firecrawl.dev", "https://firecrawl.dev/pricing"],
    prompt="Compare the features and pricing information from these pages"
)

print(result.data)

​
Job Status and Completion

Agent jobs run asynchronously. When you submit a job, you’ll receive a Job ID that you can use to check status:
  • Default method: agent() waits and returns final results
  • Start then poll: Use start_agent (Python) or startAgent (Node) to get a Job ID immediately, then poll with get_agent_status / getAgentStatus
Job results are available for 24 hours after completion.
from firecrawl import FirecrawlApp

app = FirecrawlApp(api_key="fc-YOUR_API_KEY")

# Start an agent job
agent_job = app.start_agent(
    prompt="Find the founders of Firecrawl"
)

# Check the status
status = app.get_agent_status(agent_job.id)

print(status)
# Example output:
# status='completed'
# success=True
# data={ ... }
# expires_at=datetime.datetime(...)
# credits_used=15

​
Possible States

StatusDescription
processingThe agent is still working on your request
completedExtraction finished successfully
failedAn error occurred during extraction

​
Pending Example

JSON
{
  "success": true,
  "status": "processing",
  "expiresAt": "2024-12-15T00:00:00.000Z"
}

​
Completed Example

JSON
{
  "success": true,
  "status": "completed",
  "data": {
    "founders": [
      {
        "name": "Eric Ciarla",
        "role": "Co-founder"
      },
      {
        "name": "Nicolas Camara",
        "role": "Co-founder"
      },
      {
        "name": "Caleb Peffer",
        "role": "Co-founder"
      }
    ]
  },
  "expiresAt": "2024-12-15T00:00:00.000Z",
  "creditsUsed": 15
}

​
Parameters

ParameterTypeRequiredDescription
promptstringYesNatural language description of the data you want to extract (max 10,000 characters)
urlsarrayNoOptional list of URLs to focus the extraction
schemaobjectNoOptional JSON schema for structured output

​
Agent vs Extract: What’s Improved

FeatureAgent (New)Extract
URLs RequiredNoYes
SpeedFasterStandard
CostLowerStandard
ReliabilityHigherStandard
Query FlexibilityHighModerate

​
Example Use Cases

  • Research: “Find the top 5 AI startups and their funding amounts”
  • Competitive Analysis: “Compare pricing plans between Slack and Microsoft Teams”
  • Data Gathering: “Extract contact information from company websites”
  • Content Summarization: “Summarize the latest blog posts about web scraping”

​
API Reference

Check out the Agent API Reference for more details. Have feedback or need help? Email [email protected].

​
Pricing

Firecrawl Agent uses dynamic billing that scales with the complexity of your data extraction request. You pay based on the actual work Agent performs, ensuring fair pricing whether you’re extracting simple data points or complex structured information from multiple sources.

​
How Agent pricing works

Agent pricing is dynamic and credit-based during Research Preview:
  • Simple extractions (like contact info from a single page) typically use fewer credits and cost less
  • Complex research tasks (like competitive analysis across multiple domains) use more credits but reflect the total effort involved
  • Transparent usage shows you exactly how many credits each request consumed
  • Credit conversion automatically converts agent credit usage to credits for easy billing
Credit usage varies based on the complexity of your prompt, the amount of data processed, and the structure of the output requested.

​
Getting started

All users receive 5 free daily runs to explore Agent’s capabilities without any cost. Additional usage is billed based on credit consumption and converted to credits.

​
Managing costs

Take control of your Agent spending:
  • Start with free runs: Use your 5 daily free requests to understand pricing
  • Set a maxCredits parameter: Limit your spending by setting a maximum number of credits you’re willing to spend
  • Optimize prompts: More specific prompts often use fewer credits
  • Monitor usage: Track your consumption through the dashboard
  • Set expectations: Complex multi-domain research will use more credits than simple single-page extractions
Try Agent now at firecrawl.dev/app/agent to see how credit usage scales with your specific use cases.
Pricing is subject to change as we move from Research Preview to general availability. Current users will receive advance notice of any pricing updates.