Framework Integrations

CRW integrates with popular AI agent frameworks and workflow tools. New integrations should use CRW's native /v1 API. Existing Firecrawl v2 SDK integrations can target the /firecrawl/v2 compatibility layer after validating the documented differences.

CrewAI

CrewAI support is bundled as an extra in the crw Python package — no separate package needed.

pip install "crw[crewai]"

Four ready-to-use tools — no custom BaseTool subclasses needed:

from crewai import Agent, Task, Crew
from crw.integrations.crewai import (
    CrwScrapeWebsiteTool,
    CrwCrawlWebsiteTool,
    CrwMapWebsiteTool,
    CrwSearchWebTool,
)

# Managed cloud by default (api.fastcrw.com), reads CRW_API_KEY
scrape_tool = CrwScrapeWebsiteTool()

# Or use fastCRW cloud
scrape_tool = CrwScrapeWebsiteTool(
    api_url="https://api.fastcrw.com",
    api_key="crw_live_...",
)

# Or set env vars
# export CRW_API_URL=https://api.fastcrw.com
# export CRW_API_KEY=crw_live_...

researcher = Agent(
    role="Web Researcher",
    goal="Research and summarize information from websites",
    tools=[scrape_tool],
)

task = Task(
    description="Scrape https://example.com and summarize the content",
    expected_output="A summary of the page content",
    agent=researcher,
)

crew = Crew(agents=[researcher], tasks=[task])
result = crew.kickoff()

LangChain

LangChain support ships as an extra in the crw Python package. The same loader is also published under the conventional LangChain name, so either install works and you get the identical CrwLoader.

pip install "crw[langchain]"
# or
pip install langchain-crw

from crw.integrations.langchain import CrwLoader

# Managed cloud by default (api.fastcrw.com), reads CRW_API_KEY
loader = CrwLoader(url="https://example.com", mode="scrape")
docs = loader.load()

# Cloud (fastcrw.com)
loader = CrwLoader(
    url="https://example.com",
    api_url="https://api.fastcrw.com",
    api_key="crw_live_...",
    mode="crawl",
    params={"max_depth": 3, "max_pages": 50},
)
docs = loader.load()

# Search the web (cloud only)
loader = CrwLoader(
    mode="search",
    query="web scraping tools",
    api_url="https://api.fastcrw.com",
    api_key="crw_live_...",
)
docs = loader.load()

Flowise (PR pending)

CRW node submitted: PR #6066

Provides Scrape, Crawl, and Map nodes that connect to your CRW instance. Configure the CRW API URL in the node settings.

Agno (PR pending)

CRW toolkit submitted: PR #7183

Once merged, usage will be:

from agno.agent import Agent
from agno.tools.crw import CrwTools

agent = Agent(tools=[CrwTools()])
agent.print_response("Scrape https://example.com and summarize it")

OpenClaw

OpenClaw integrates with CRW via the MCP server — no separate plugin package is needed. Point OpenClaw at the crw-mcp binary and it gains Scrape, Crawl, Map, and Search tools automatically.

See the MCP documentation for configuration details.

n8n

Published npm package: n8n-nodes-crw

Install via n8n UI: Settings > Community Nodes > Install > n8n-nodes-crw

Or via Docker:

docker run -e EXTRA_COMMUNITY_PACKAGES=n8n-nodes-crw n8nio/n8n

Source: github.com/us/n8n-nodes-crw

MCP (10+ Platforms)

CRW includes a built-in MCP server that works with any MCP-compatible platform. See the MCP documentation for setup instructions for:

• Claude Code, Claude Desktop
• Cursor, Windsurf
• Cline, Continue.dev
• OpenClaw
• OpenAI Codex CLI
• Gemini CLI
• VS Code GitHub Copilot Agent
• Roo Code

Automation Tools

Firecrawl v2 SDK Migration

For existing Firecrawl v2 SDK projects, point the SDK at a CRW engine and validate the compatibility matrix before switching production traffic:

# Point the Firecrawl SDK at your CRW instance
export FIRECRAWL_API_URL=http://localhost:3000

from firecrawl import FirecrawlApp

# This talks to CRW, not Firecrawl
app = FirecrawlApp(api_url="http://localhost:3000")
result = app.scrape_url("https://example.com")

Note: This is a migration path, not the default recommendation for new CRW code. New integrations should use the CRW SDKs or direct /v1 HTTP routes.

Python (Direct HTTP)

import requests

response = requests.post("https://api.fastcrw.com/v1/scrape", json={
    "url": "https://example.com",
    "formats": ["markdown", "links"]
})
data = response.json()["data"]
print(data["markdown"])

Node.js (Direct HTTP)

const response = await fetch("https://api.fastcrw.com/v1/scrape", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    url: "https://example.com",
    formats: ["markdown", "links"]
  })
});
const { data } = await response.json();
console.log(data.markdown);

Building Custom Integrations

The API is straightforward enough that most integrations are a thin wrapper:

1. Set the Authorization header with your API key.
2. POST JSON to the endpoint you need (/v1/scrape, /v1/crawl, /v1/map, /v1/search).
3. Parse the JSON response.

No SDK is required — the consistent API design means any HTTP client works.

Choosing Between Cloud and Self-Hosted

Both options expose the same API, so your integration code works with either.

Endpoint Support Matrix

Not every integration supports every endpoint. Search requires a cloud API backend or a search sidecar configured for the embedded MCP server.

All Integrations