# How to connect Crawl4AI to Claude

```json
{
  "title": "How to connect Crawl4AI to Claude",
  "toolkit": "Crawl4AI",
  "toolkit_slug": "crawl4ai",
  "framework": "Claude Cowork",
  "framework_slug": "claude-cowork",
  "url": "https://composio.dev/toolkits/crawl4ai/framework/claude-cowork",
  "markdown_url": "https://composio.dev/toolkits/crawl4ai/framework/claude-cowork.md",
  "updated_at": "2026-11-17T05:35:02.235Z"
}
```

## Introduction

Claude is Anthropic's AI assistant, available on the web, in the desktop app, and on mobile. Cowork is its agent for knowledge work, and it runs on all of those surfaces too. Connected to your apps, Claude can work with your files and services to accomplish complex tasks on your behalf.
This guide walks you through the easiest and most secure way to connect your Crawl4AI account to Claude via Composio Connect, enabling it to search web for competitor pricing pages, extract product data from retailer pages, verify email addresses from lead lists, and more such actions on your behalf without compromising your account security.
Setup is the same on Claude Web, Desktop, and Cowork, and you only need to do it once. The connector is tied to your account, so it's available in all three.
For exact tool slugs and parameter tables, see the [tool catalog in the Composio docs](https://docs.composio.dev/toolkits/crawl4ai).

## Also integrate Crawl4AI with

- [ChatGPT](https://composio.dev/toolkits/crawl4ai/framework/chatgpt)
- [Hermes](https://composio.dev/toolkits/crawl4ai/framework/hermes-agent)
- [Atomic Agent](https://composio.dev/toolkits/crawl4ai/framework/atomic-agent)

## Connect Crawl4AI to Claude Cowork

### Connecting Crawl4AI to Claude Web, Desktop, and Cowork
Two ways to add the Composio connector. Pick whichever you prefer.
### 1. Add with one click
Click the button. Claude opens with the Composio connector filled in. Click Add, then go to step 2.
[Add to Claude](https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=Composio&connectorUrl=https%3A%2F%2Fconnect.composio.dev%2Fmcp)
### Add manually
In Claude Desktop, click Customize in the left sidebar, then select Connectors and click the + icon at the top.
Click Add custom connector and paste in the Composio MCP server URL:

```bash
https://connect.composio.dev/mcp
```

## What is the Crawl4AI MCP server, and what's possible with it?

The Crawl4AI MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Crawl4AI account. It provides structured and secure access so your agent can perform Crawl4AI operations on your behalf.

## Supported Tools

| Tool slug | Name | Description |
|---|---|---|
| `CRAWL4AI_ANSWER_QUESTION` | Answer Question | Answer a question from web sources and return the answer with citations when available. This Crawl4AI endpoint is experimental. |
| `CRAWL4AI_CANCEL_RECIPE_JOB` | Cancel Recipe Job | Stop a running recipe job. The provider retains it in paused state and may report a pause reason and partial usage. |
| `CRAWL4AI_CANCEL_SITE_CRAWL` | Cancel Site Crawl | Stop a running site crawl job; the job ends in the paused state. |
| `CRAWL4AI_CONSUME_MONITOR_CHANGES` | Read Monitor Changes | Read the oldest new changes across monitors and mark the returned feed items as read. Without since, later calls start after consumed items; pass a prior cursor to replay. Requires confirm_mark_read=true; use agent to limit which agent's feed is consumed. |
| `CRAWL4AI_CREATE_MONITOR` | Create Monitor | Create a page-change monitor, or return the existing equivalent monitor, that performs recurring credit-consuming checks until paused or deleted and can send configured notifications. |
| `CRAWL4AI_CREATE_PLAN_CHECKOUT` | Create Plan Checkout | Create an account-specific Stripe checkout flow for a Crawl4AI paid plan. This only returns a checkout URL: it does not directly charge a card or start the plan. Open and pay the Stripe checkout URL to complete the subscription. Use GET_PRICES_AND_PLANS first for current plan names and prices. |
| `CRAWL4AI_DELETE_MONITOR` | Delete Monitor | Permanently delete one page-change monitor and its history. Use this for intentional removal and test cleanup. |
| `CRAWL4AI_ESTIMATE_COST` | Estimate Cost | Estimate the minimum and maximum Crawl4AI credits for a proposed API request without running that request. |
| `CRAWL4AI_EXTRACT_PRICES` | Extract Prices | Read pricing pages for up to 20 site-and-country combinations and return documented plans, billing periods, add-ons, discounts, trials, and source quotes. |
| `CRAWL4AI_EXTRACT_STRUCTURED_DATA` | Extract Structured Data | Extract structured records from exactly one URL or supplied content using an instruction, a JSON Schema, or both. |
| `CRAWL4AI_FORGET_EMAIL_VERIFICATION` | Forget Email Verification | Request permanent removal of Crawl4AI's stored verification answer for one email. This destructive privacy operation is documented as returning 204, but the provider may currently reject it with HTTP 405. |
| `CRAWL4AI_GET_BILLING_BALANCE` | Get Billing Balance | Return the connected account's current free-credit balance, allowance, plan, tier, card status, spending, and top-up balance. Ledger values can trail the exact response-header balance by about one minute. |
| `CRAWL4AI_GET_COMPANY` | Get Lead Company | Get one private-preview Lead company by domain, LinkedIn page, or an unambiguous name optionally narrowed by country. |
| `CRAWL4AI_GET_EMAIL_VERIFICATION_JOB` | Get Email Verification Job | Get later results for a batch email verification job, including background retries. |
| `CRAWL4AI_GET_MONITOR` | Get Monitor | Return one page-change monitor's configuration, health, baseline, cost, and notification settings. |
| `CRAWL4AI_GET_MONITOR_VERSION` | Get Monitor Version | Download the page exactly as Crawl4AI captured it for monitor evidence, returned as a Markdown text file. |
| `CRAWL4AI_GET_PERSON` | Get Lead Person | Get one private-preview Lead person and held contact data by one documented lookup. |
| `CRAWL4AI_GET_PERSON_EMAIL` | Get Lead Person Email | Get private-preview personal contact data: one person's checked email by one documented lookup. Lead may construct and verify a likely address from the company pattern before answering. |
| `CRAWL4AI_GET_PERSON_PHONE` | Get Lead Person Phone | Get private-preview personal contact data: a person's direct and company phones by one documented lookup. |
| `CRAWL4AI_GET_PRICE_HISTORY` | Get Price History | Return up to 1,000 observed price changes for one domain and optional country, newest first. |
| `CRAWL4AI_GET_PRICES_AND_PLANS` | Get Prices and Plans | Return Crawl4AI's current credit prices, effort levels, plans, allowances, and request limits from the authoritative live catalog. |
| `CRAWL4AI_GET_RECIPE_HEALTH` | Get Recipe Health | Get the latest regional health check and release stage for every recipe, keyed by its live catalog name. |
| `CRAWL4AI_GET_RECIPE_JOB` | Get Recipe Job | Get a recipe job's status, progress counts, pause reason, and usage. |
| `CRAWL4AI_GET_RECIPE_JOB_RESULTS` | Get Recipe Job Results | Download a recipe job's result rows as an NDJSON file. Each line is one dynamic recipe output row with a sequence number. |
| `CRAWL4AI_GET_SCRAPE_JOB` | Get Scrape Job | Return a scrape job's current status, URL counts, and credit usage, with optional per-URL detail. |
| `CRAWL4AI_GET_SCRAPE_JOB_RESULTS` | Get Scrape Job Results | Download one page of up to 500 scrape job results as NDJSON; pass the next result offset explicitly to retrieve another page. |
| `CRAWL4AI_GET_SERP` | Get SERP | Return structured Google rankings through Crawl4AI's GET transport, including source selection and the fixed device, engine, and result-type controls. Use SEARCH_SERP for the streamlined POST request. |
| `CRAWL4AI_GET_SITE_CRAWL_JOB` | Get Site Crawl Job | Read a site crawl job's state, progress counts, completion deadline, usage, and pause reason when applicable. |
| `CRAWL4AI_GET_SITE_CRAWL_RESULTS` | Get Site Crawl Results | Download one page of up to 500 site crawl result rows as NDJSON, optionally continuing after a previously returned sequence number. |
| `CRAWL4AI_LIST_BILLING_TOPUPS` | List Billing Topups | List account payments and credit additions newest first, including receipt URLs, one cursor-controlled page at a time. |
| `CRAWL4AI_LIST_LEAD_INDUSTRIES` | List Lead Industries | Return documented LinkedIn industry names or names closest to query words. |
| `CRAWL4AI_LIST_MONITOR_EVENTS` | List Monitor Events | List one monitor's change and health timeline oldest first, including facts and links to captured evidence versions. This does not mark the feed read. |
| `CRAWL4AI_LIST_MONITORS` | List Monitors | List the connected account's page-change monitors, optionally filtered by agent label. |
| `CRAWL4AI_LIST_RECIPES` | List Recipes | List the live ready-made scraper catalog, including each recipe's dynamic input schema, output fields, cost, stage, and health. |
| `CRAWL4AI_PURCHASE_EXTRA_CREDITS` | Purchase Extra Credits | Immediately charge the saved card for the connected account's paid Crawl4AI plan and add non-expiring extra credits. This is a real, irreversible card charge, not a checkout flow. Call only after the user explicitly approves the exact US-dollar amount and set confirm_charge=true. |
| `CRAWL4AI_RETRY_SCRAPE_JOB` | Retry Scrape Job | Re-run only the failed URLs in an existing scrape job. |
| `CRAWL4AI_RUN_RECIPE` | Run Recipe | Run one synchronous catalog recipe with inputs that match its current schema. Job-only recipes and catalog inputs marked secret are rejected. |
| `CRAWL4AI_SCAN_SITE_URLS` | Scan Site URLs | Discover a site's known public page URLs from sitemaps, its home page, Common Crawl, and optionally the Wayback Machine without crawling each URL. |
| `CRAWL4AI_SCRAPE_PAGE` | Scrape Page | Fetch one public web page and return its Markdown, HTML, and optional parsed links, media, metadata, or tables. |
| `CRAWL4AI_SCRAPE_PAGES` | Scrape Pages | Scrape up to 50 public URLs with shared options and return a downloadable NDJSON file of completion rows followed by a final summary row. |
| `CRAWL4AI_SEARCH_COMPANIES` | Search Lead Companies | Search private-preview Lead companies by documented company filters and optional description-ranking words, one provider-cursor page at a time. |
| `CRAWL4AI_SEARCH_PEOPLE` | Search Lead People | Search private-preview Lead people by documented company and person filters. Results can contain personal contact data; people who exercised removal are excluded permanently. Returns one provider-cursor page at a time. |
| `CRAWL4AI_SEARCH_SERP` | Search SERP | Return structured Google result ranks and SERP blocks for a localized query and explicit result page. |
| `CRAWL4AI_SEARCH_WEB` | Search Web | Search the web for ranked, cleaned results, optionally including related questions, queries, news, videos, and discussions. |
| `CRAWL4AI_SUBMIT_RECIPE_JOB` | Submit Recipe Job | Submit an asynchronous run for a job-capable recipe using its dynamic live catalog inputs. For execution-log safety, catalog inputs marked secret are intentionally unsupported and rejected before submission. |
| `CRAWL4AI_SUBMIT_SCRAPE_JOB` | Submit Scrape Job | Submit up to 10,000 public URLs for asynchronous scraping and return a job identifier for status polling and paged result retrieval. |
| `CRAWL4AI_SUBMIT_SITE_CRAWL` | Submit Site Crawl | Start a background crawl of public pages within one site and return the job identifier used to inspect its status and results. |
| `CRAWL4AI_UPDATE_MONITOR` | Update Monitor | Change a monitor's watch rule, schedule, notifications, schema, name, or paused state. At least one change must be supplied. |
| `CRAWL4AI_VERIFY_EMAIL` | Verify Email | Check whether one email can receive mail. Crawl4AI performs an SMTP conversation but sends no mail; stored answers may be reused for 30 days. |
| `CRAWL4AI_VERIFY_EMAIL_DOMAIN` | Verify Email Domain | Check one domain's mail acceptance, provider, gateway, catch-all, SPF, and DMARC results. |
| `CRAWL4AI_VERIFY_EMAILS` | Verify Emails | Check one to 1,000 email addresses without sending mail. Greylisted addresses return unknown initially and are retried in the background. |

## Supported Triggers

None listed.

## Troubleshooting

### 1. Why does Claude not use the Composio connector for Crawl4AI?

The connector can show as connected, but Claude must use it in each Crawl4AI conversation.
- Click + at the lower left of the chat, or type /.
- Hover Connectors.
- Turn on Composio.
- Ask Claude to connect to Crawl4AI again.

### 2. Why does Claude show an MCP server error before Crawl4AI connects?

Claude can show Couldn't reach the MCP server or Authorization with the MCP server failed when you add the connector for Crawl4AI.
- Check that the connector URL is https://connect.composio.dev/mcp.
- Connectors run from Anthropic's cloud, so your device network is usually not the problem.
- Add the connector again from Customize > Connectors, then ask Claude to connect Crawl4AI.
- If it fails again, send the error and the ofid_ reference id to support@composio.dev.

### 3. Why does the Crawl4AI authorization link not work?

The Composio connector is one server. Your Crawl4AI account connects separately when Claude asks for it.
- If the browser does not open, or the link expires, ask Claude to retry the Crawl4AI connection.
- Composio creates a fresh Crawl4AI authorization link.
- Open that link in your browser and finish the Crawl4AI sign-in.

### 4. Why do Crawl4AI actions fail after they worked before?

The Crawl4AI connection can expire or lose permission.
- Ask Claude to inspect the Crawl4AI connection.
- If the connection is unhealthy, disconnect it when Claude prompts you.
- Reconnect the Crawl4AI account and retry the action.
- If Claude shows a permission prompt, approve it before Claude changes data in Crawl4AI.
- If the action still fails with a permission error, check that the connected account has that permission in Crawl4AI itself.

### 5. How do I switch to a different Crawl4AI account?

Claude can manage the Crawl4AI connection for you.
- Ask Claude to disconnect or delete the current Crawl4AI account.
- Ask Claude to connect Crawl4AI again.
- Open the new Composio authorization link for Crawl4AI.
- Sign in with the other Crawl4AI account in the browser.
- After the account connects, ask Claude to retry the Crawl4AI task.

### 6. Why is my Crawl4AI connection missing in Claude Desktop or on a Team plan?

The connector belongs to your Claude account, so your Crawl4AI connection works in Claude Web, Desktop, and Cowork.
- In Claude Desktop, open Customize > Connectors.
- Next to Composio, click ⋮ and click Clear cache.
- If Crawl4AI tools still do not appear, click ⋮, disconnect, remove the connector, and add it again.
- On Team or Enterprise, an Owner or Primary Owner must enable the connector before you can connect Crawl4AI.

## Creating MCP Server - Stand-alone vs Composio SDK

The Crawl4AI MCP server connects Claude Web, Desktop, and Cowork to your Crawl4AI account through Composio. Once connected, Claude can use the available Crawl4AI tools and triggers to complete tasks on your behalf.

## Complete Code

None listed.

## How to build Crawl4AI MCP Agent with another framework

- [ChatGPT](https://composio.dev/toolkits/crawl4ai/framework/chatgpt)
- [Hermes](https://composio.dev/toolkits/crawl4ai/framework/hermes-agent)
- [Atomic Agent](https://composio.dev/toolkits/crawl4ai/framework/atomic-agent)

## Related Toolkits

- [Supabase](https://composio.dev/toolkits/supabase) - Supabase is an open-source backend platform offering scalable Postgres databases, authentication, storage, and real-time APIs. It lets developers build modern apps without managing infrastructure.
- [Codeinterpreter](https://composio.dev/toolkits/codeinterpreter) - Codeinterpreter is a Python-based coding environment with built-in data analysis and visualization. It lets you instantly run scripts, plot results, and prototype solutions inside supported platforms.
- [GitHub](https://composio.dev/toolkits/github) - GitHub is a code hosting platform for version control and collaborative software development. It streamlines project management, code review, and team workflows in one place.
- [1password](https://composio.dev/toolkits/_1password) - 1Password is a password manager and digital vault for storing logins, secrets, notes, and secure documents. It helps individuals and teams protect credentials, share access safely, and reduce password risk.
- [Ably](https://composio.dev/toolkits/ably) - Ably is a real-time messaging platform for live chat and data sync in modern apps. It offers global scale and rock-solid reliability for seamless, instant experiences.
- [Abuselpdb](https://composio.dev/toolkits/abuselpdb) - Abuselpdb is a central database for reporting and checking IPs linked to malicious online activity. Use it to quickly identify and report suspicious or abusive IP addresses.
- [Alchemy](https://composio.dev/toolkits/alchemy) - Alchemy is a blockchain development platform offering APIs and tools for Ethereum apps. It simplifies building and scaling Web3 projects with robust infrastructure.
- [Algolia](https://composio.dev/toolkits/algolia) - Algolia is a hosted search API that powers lightning-fast, relevant search experiences for web and mobile apps. It helps developers deliver instant, typo-tolerant, and scalable search without complex infrastructure.
- [Anakin](https://composio.dev/toolkits/anakin_io) - Anakin is a web automation platform offering scraping, crawling, search, and browser automation. Use it to extract structured data, automate site tasks, and monitor web changes reliably.
- [Anchor browser](https://composio.dev/toolkits/anchor_browser) - Anchor browser is a developer platform for AI-powered web automation. It transforms complex browser actions into easy API endpoints for streamlined web interaction.
- [AnyAPI](https://composio.dev/toolkits/anyapi) - AnyAPI is a marketplace of utility APIs for validation, currency conversion, holidays, documents, and web data. It brings common application workflows together through one practical API platform.
- [Apiflash](https://composio.dev/toolkits/apiflash) - Apiflash is a website screenshot API for programmatically capturing web pages. It delivers high-quality screenshots on demand for automation, monitoring, or reporting.
- [Apiverve](https://composio.dev/toolkits/apiverve) - Apiverve delivers a suite of powerful APIs that simplify integration for developers. It's designed for reliability and scalability so you can build faster, smarter applications without the integration headache.
- [Appcircle](https://composio.dev/toolkits/appcircle) - Appcircle is an enterprise-grade mobile CI/CD platform for building, testing, and publishing mobile apps. It streamlines mobile DevOps so teams ship faster and with more confidence.
- [Appdrag](https://composio.dev/toolkits/appdrag) - Appdrag is a cloud platform for building websites, APIs, and databases with drag-and-drop tools and code editing. It accelerates development and iteration by combining hosting, database management, and low-code features in one place.
- [Appveyor](https://composio.dev/toolkits/appveyor) - AppVeyor is a cloud-based continuous integration service for building, testing, and deploying applications. It helps developers automate and streamline their software delivery pipelines.
- [Authyo](https://composio.dev/toolkits/authyo) - Authyo is a REST API service for passwordless authentication, OTP verification, session management, and transactional notifications. Use it to add secure login flows and user verification without building auth infrastructure from scratch.
- [AWS Marketplace MCP](https://composio.dev/toolkits/aws_marketplace_mcp) - AWS Marketplace MCP provides MCP access to AWS Marketplace's cloud software, data, and services catalog. Use it to discover, compare, and evaluate 30K+ AWS Marketplace listings faster.
- [Azure Monitor Activity Log](https://composio.dev/toolkits/azure_monitor_activity_log) - Azure Monitor Activity Log is Azure's service that records management and control-plane events for a subscription. Use it to audit changes, troubleshoot issues, and track resource operations across your Azure resources.
- [Backendless](https://composio.dev/toolkits/backendless) - Backendless is a backend-as-a-service platform for mobile and web apps, offering database, file storage, user authentication, and APIs. It helps developers ship scalable applications faster without managing server infrastructure.

## Frequently Asked Questions

### What are the differences in Tool Router MCP and Crawl4AI MCP?

With a standalone Crawl4AI MCP server, the agents and LLMs can only access a fixed set of Crawl4AI tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Crawl4AI and many other apps based on the task at hand, all through a single MCP endpoint.

### Can I use Tool Router MCP with Claude Cowork?

Yes, you can. Claude Cowork fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Crawl4AI tools.

### Can I manage the permissions and scopes for Crawl4AI while using Tool Router?

Yes, absolutely. You can configure which Crawl4AI scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

### How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Crawl4AI data and credentials are handled as safely as possible.

---
[See all toolkits](https://composio.dev/toolkits) · [Composio docs](https://docs.composio.dev/llms.txt)
