How to Give ChatGPT Web Scraping with MCP Connectors (2026)

ChatGPT can now call your own tools through custom MCP connectors -- including web scraping. But there is a catch the marketing pages skip: connectors must be remote servers, so a local tool like CrawlForge cannot be pasted in directly. This guide is the honest version: what is actually possible, why a wrapper is needed, and the exact bridge to build.

Table of Contents

• What ChatGPT Connectors Are
• Which Plans Can Use Them
• The Transport Catch: Remote Only
• Why CrawlForge Needs a Wrapper
• Build the Bridge: A Remote MCP Wrapper
• Add the Connector in ChatGPT
• Auth and Safety
• A Simpler Alternative

What ChatGPT Connectors Are

ChatGPT supports custom MCP connectors -- renamed "apps" in December 2025, so the current UI says "Apps & Connectors." Through Developer mode, you connect an external MCP server and ChatGPT can call its tools mid-conversation, asking you to confirm before any write action. It is the same Model Context Protocol that powers web scraping in Claude -- just a different client.

Developer mode (full read/write MCP) is explicitly a beta; OpenAI notes the UI and permissions may change.

Which Plans Can Use Them

Per OpenAI's plan table, the ability to add a custom MCP connector is available on:

• Plus, Pro, Business, Enterprise, and Edu -- yes.
• Free and Go -- no.

Full write-action support is rolling out most broadly to Business, Enterprise, and Edu. If you only need ChatGPT to read scraped data, the read-only path below is enough.

The Transport Catch: Remote Only

This is the part that trips people up. A ChatGPT connector must be a remote MCP server reachable over HTTPS, using SSE or Streamable HTTP transport. You paste a URL into ChatGPT; you do not point it at a command on your machine.

That rules out local stdio servers -- the kind you install with npx. To use one, you either host it publicly or expose a local server through a tunnel such as ngrok or Cloudflare Tunnel.

There is also a tool-naming rule worth knowing: ChatGPT's deep research and company-knowledge paths require your server to expose two read-only tools named search and fetch with a specific schema. Full Developer mode lets you expose arbitrary tools, so that two-tool constraint applies only to the deep-research path.

Why CrawlForge Needs a Wrapper

CrawlForge ships as a local stdio MCP server (via npx) plus a REST API at https://www.crawlforge.dev/api/v1/tools/. Neither is a remote MCP URL, and its tools are named search_web, fetch_url, and extract_content -- not the search/fetch pair ChatGPT's deep-research mode expects.

So you cannot paste CrawlForge straight into ChatGPT today. The practical path is a thin remote MCP wrapper: a small server you host that exposes ChatGPT-shaped tools and calls CrawlForge's REST API behind the scenes. It is about 30 lines.

Build the Bridge: A Remote MCP Wrapper

FastMCP (Python) is the quickest way to stand up a remote MCP server with the search and fetch tools ChatGPT wants. Each tool calls CrawlForge's REST API using your cf_live_ key in the X-API-Key header:

# server.py
import os
import httpx
from fastmcp import FastMCP

mcp = FastMCP("CrawlForge Bridge")
BASE = "https://www.crawlforge.dev/api/v1/tools"
HEADERS = {"X-API-Key": os.environ["CRAWLFORGE_API_KEY"]}

@mcp.tool()
async def search(query: str) -> list[dict]:
    """Search the web. Returns id/title/url results for ChatGPT."""
    async with httpx.AsyncClient(timeout=30) as client:
        r = await client.post(f"{BASE}/search_web", headers=HEADERS,
                              json={"query": query, "limit": 10})
    results = r.json().get("results", [])
    return [{"id": x["link"], "title": x["title"], "url": x["link"]} for x in results]

@mcp.tool()
async def fetch(id: str) -> dict:
    """Fetch full page content by id (the URL) for ChatGPT."""
    async with httpx.AsyncClient(timeout=30) as client:
        r = await client.post(f"{BASE}/extract_content", headers=HEADERS,
                              json={"url": id})
    data = r.json()
    return {"id": id, "title": data.get("title", id),
            "text": data.get("content", ""), "url": id}

if __name__ == "__main__":
    mcp.run(transport="http", host="0.0.0.0", port=8000)

Run it, then expose it over HTTPS. For a quick test, tunnel your local port:

pip install fastmcp httpx
export CRAWLFORGE_API_KEY="cf_live_your_key_here"
python server.py
# in another terminal:
ngrok http 8000

For Developer mode you can skip the search/fetch naming and instead map tools one-to-one to CrawlForge -- expose scrape_structured, stealth_mode, or deep_research directly. The wrapper pattern is the same.

Add the Connector in ChatGPT

1. Open Settings -> Apps & Connectors -> Advanced and enable Developer mode.
2. Go to Apps & Connectors -> Create.
3. Paste your public HTTPS MCP URL (for example, your ngrok URL plus /mcp), name it, and choose an auth method.
4. Confirm the "I trust this application" checkbox.

Your search and fetch tools now appear. In a chat, select the connector and ask ChatGPT to research a topic -- it will call search, then fetch the most relevant results through CrawlForge.

Auth and Safety

Connectors authenticate with none (public) or OAuth -- there is no simple API-key-header option in the ChatGPT UI, which is exactly why the wrapper holds your CrawlForge key server-side. ChatGPT also asks you to confirm before write actions, and you can inspect each tool call's payload before approving it.

Take OpenAI's warnings seriously: only connect servers you trust. A custom connector increases risk, including prompt injection, and a model mistake on a write action could destroy or leak data. A read-only scraping bridge like the one above is low-risk; lock it down with OAuth before sharing it.

A Simpler Alternative

If you would rather not host anything, use CrawlForge from code with the OpenAI Agents SDK or Responses API -- no remote MCP server required. See CrawlForge with the OpenAI Agents SDK for that path, and the complete guide to MCP web scraping for the bigger picture.

Get a free CrawlForge API key -- 1,000 credits, no credit card required, and wire it into your bridge in minutes.