We raised a $50M Series B led by BenchmarkLearn more
Cloudflare logo — MCP server on Gumloop

Cloudflare MCP Server

Connect to the Cloudflare MCP server to search documentation, find the right API endpoints, and manage DNS, Workers, and account settings using AI agents on Gumloop, Claude, Cursor, or Codex.

Book a demo

Try Cloudflare now

Type what you want done. Sign in and run it live with an AI agent.

|

Cloudflare logoCloudflare
Gradient

Installation

Set up the Cloudflare MCP server in Gumloop

Do this once to provision your hosted server URL.

1

Create a Gumloop account

To use this MCP, you need a Gumloop account. If you don't have one yet, sign up and start a 14-day free trial.

2

Add and authorize the Cloudflare server

In Gumloop, open Connectors and add Cloudflare. You'll be sent to Cloudflare's sign-in screen to grant access. The credential is stored securely in Gumloop.

Then use it in your client

Use in GumloopUse in Gumloop
1

Use Cloudflare in an agent

Once Cloudflare is set up, just open any Gumloop agent, add Cloudflare as a connector, and start chatting with the agent.

Tools (3)

  • Docs

    Search the Cloudflare documentation. This tool should be used to answer any question about Cloudflare products or features, including: - Workers, Pages, R2, Images, Stream, D1, Durable Objects, KV, Workflows, Hyperdrive, Queues - AI Search, Workers AI, Vectorize, AI Gateway, Browser Rendering - Zero Trust, Access, Tunnel, Gateway, Browser Isolation, WARP, DDOS, Magic Transit, Magic WAN - CDN, Cache, DNS, Zaraz, Argo, Rulesets, Terraform, Account and Billing Results are returned as semantically similar chunks to the query.

  • Search

    Search the Cloudflare OpenAPI spec. All $refs are pre-resolved inline. Products: ai, cloudforce-one, access, magic, workers, dlp, realtime, devices, email-security, ai-gateway, gateway, intel, email, stream, ai-search, brand-protection, dex, api_gateway, browser-rendering, builds, addressing, rulesets, security-center, settings, logpush, vectorize, firewall, load_balancers, logs, artifacts... (146 total) Types: interface OperationInfo { summary?: string; description?: string; tags?: string[]; parameters?: Array<{ name: string; in: string; required?: boolean; schema?: unknown; description?: string }>; requestBody?: { required?: boolean; content?: Record<string, { schema?: unknown }> }; responses?: Record<string, { description?: string; content?: Record<string, { schema?: unknown }> }>; } interface PathItem { get?: OperationInfo; post?: OperationInfo; put?: OperationInfo; patch?: OperationInfo; delete?: OperationInfo; } declare const spec: { paths: Record<string, PathItem>; }; Examples: // Find endpoints by product async () => { const results = []; for (const [path, methods] of Object.entries(spec.paths)) { for (const [method, op] of Object.entries(methods)) { if (op.tags?.some(t => t.toLowerCase() === 'workers')) { results.push({ method: method.toUpperCase(), path, summary: op.summary }); } } } return results; } // Get endpoint with requestBody schema (refs are resolved) async () => { const op = spec.paths['/accounts/{account_id}/d1/database']?.post; return { summary: op?.summary, requestBody: op?.requestBody }; } // Get endpoint parameters async () => { const op = spec.paths['/accounts/{account_id}/workers/scripts']?.get; return op?.parameters; }

  • Execute

    Execute JavaScript code against the Cloudflare API. First use the 'search' tool to find the right endpoints, then write code using the cloudflare.request() function. Available in your code: interface CloudflareRequestOptions { method: "GET" | "POST" | "PUT" | "PATCH" | "DELETE"; path: string; query?: Record<string, string | number | boolean | undefined>; body?: unknown; contentType?: string; // Custom Content-Type header (defaults to application/json if body is present) rawBody?: boolean; // If true, sends body as-is without JSON.stringify } interface CloudflareResponse<T = unknown> { success: boolean; status: number; result: T; errors: Array<{ code: number; message: string }>; messages: Array<{ code: number; message: string }>; result_info?: { page: number; per_page: number; total_pages: number; count: number; total_count: number; }; } declare const cloudflare: { request<T = unknown>(options: CloudflareRequestOptions): Promise<CloudflareResponse<T>>; }; declare const accountId: string; // accountId is pre-set to "b968f3db90c2b97603172e1d528b0538" (Revecitycrypto@gmail.com's Account) — use it directly in API paths. Your code must be an async arrow function that returns the result. Example: Worker with bindings (requires multipart/form-data): async () => { const code = `addEventListener('fetch', e => e.respondWith(MY_KV.get('key').then(v => new Response(v || 'none'))));`; const metadata = { body_part: "script", bindings: [{ type: "kv_namespace", name: "MY_KV", namespace_id: "your-kv-id" }] }; const b = `--F${Date.now()}`; const body = [`--${b}`, 'Content-Disposition: form-data; name="metadata"', 'Content-Type: application/json', '', JSON.stringify(metadata), `--${b}`, 'Content-Disposition: form-data; name="script"', 'Content-Type: application/javascript', '', code, `--${b}--`].join("\r\n"); return cloudflare.request({ method: "PUT", path: `/accounts/${accountId}/workers/scripts/my-worker`, body, contentType: `multipart/form-data; boundary=${b}`, rawBody: true }); }

What is Cloudflare MCP?

The Cloudflare MCP server gives AI agents access to your Cloudflare account and the full Cloudflare API. That means agents can search Cloudflare’s product documentation, look up the exact endpoint for a task in Cloudflare’s OpenAPI spec, and run code against that API to read or change account resources. It works across Cloudflare’s product surface, including DNS, Workers, zones, and account and security settings.

If you manage Cloudflare by clicking through the dashboard or hand-writing scripts against the API, you might be spending time manually looking up the right endpoints in the docs, wiring up authentication, handling pagination, and parsing JSON responses. Instead, try an AI agent: describe the change you want, and your AI agent will find the endpoint and run the API call for you.

MCP stands for Model Context Protocol. It’s an open standard that gives AI agents a way to connect to external tools and services. Instead of managing Cloudflare API tokens, wiring up authentication, and writing code against the REST API yourself, you just connect your Cloudflare account to Gumloop once. After that, you can manage DNS, Workers, and account settings just by chatting with your AI agent.

Related MCP servers

What you can do with Cloudflare MCP on Gumloop

  • Search the Cloudflare documentation

    Ask questions about any Cloudflare product or feature and get answers pulled straight from the official docs. Your AI agent works from current, accurate information instead of guesswork.

  • Find the right API endpoint fast

    Search Cloudflare’s OpenAPI spec, with all references resolved inline, to locate the exact endpoint and parameters for a task across products like Workers, Access, Magic Transit, DLP, and Realtime.

  • Run API calls against your account

    Execute code against the Cloudflare API to read or change account resources. Your agent can chain the search step and the execute step so a plain request turns into a real API action.

  • Manage DNS records

    Read, create, update, and delete DNS records across your zones through the API, so an agent can add a record, fix a misconfiguration, or audit your DNS without you touching the dashboard.

  • Inspect and configure Workers

    Check what’s deployed and update Cloudflare Workers and their settings programmatically. Useful for shipping edge logic or reviewing configuration.

  • Read and adjust zone and account settings

    Pull zone configuration, security settings, and account details, then apply changes through the API when you need them.

How to connect the Gumloop Cloudflare MCP Server

  1. 1

    Create a Gumloop account

    Sign up at gumloop.com. Every new account starts with a 14-day free trial.

  2. 2

    Add the Cloudflare MCP server

    Copy your MCP server URL from Gumloop and add it to your preferred client (Claude, Cursor, or Gumloop workflows). You'll authorize on first use.

  3. 3

    Start using Cloudflare in your AI workflows

    That's it. That’s it: your hosted Cloudflare MCP server is live, and the same server URL works in any MCP client you’ve added it to. Your AI agent can now search Cloudflare’s docs, find the right endpoints, and run API calls to manage DNS, Workers, and account settings. Use it inside a Gumloop automation, in Claude Desktop, or in Cursor.

Cloudflare MCP use cases

On-demand DNS management for platform teams

When a new service needs a subdomain, a Gumloop agent can look up the right DNS endpoint, create the record in the correct zone, verify that it resolved, and post a confirmation to Slack. Platform teams no longer have to hand-edit records in the dashboard for every request.

Incident response for on-call engineers

During an incident, an on-call engineer can ask the agent to pull the current configuration for a zone or Worker, check recent settings, and apply a fix such as toggling a security setting. And they can do it all from the chat window, rather than digging through the dashboard.

Documentation-aware answers for developer experience teams

An agent can search Cloudflare’s docs to answer a product question, then confirm the exact API call in the OpenAPI spec. Internal questions get accurate answers with the endpoint to back them up.

Automated configuration audits for security teams

On a schedule, an agent can read zone and account settings across your Cloudflare account, compare them against your baseline, and write a report to Google Sheets or flag drift in Slack. Security teams can catch misconfigurations without a manual review.

Cross-tool infrastructure agents

Connect Cloudflare with GitHub, Slack, and Datadog in a single agent. When a deploy lands, the agent can update a Worker, confirm the change, and post the result to your team channel, so your edge platform talks to the rest of your stack.

Why use Gumloop for Cloudflare MCP

  • Connect once, no token juggling

    Most Cloudflare MCP setups you find on GitHub ask you to create and store an API token, set environment variables, and write code to handle authentication and the REST API. With Gumloop, you connect your Cloudflare account once and the connection is handled for you. No config files, token juggling, or coding against the Cloudflare API necessary.

  • Works with multiple MCP clients

    Use the Cloudflare MCP server in Claude Desktop, Cursor, Codex, or directly inside Gumloop agents. Same server URL, works with any MCP client.

  • Chain Cloudflare with 100+ other integrations

    Combine Cloudflare with GitHub, Slack, Datadog, Google Sheets, and other MCP tools in a single AI agent. Pull data from one tool, process it, and write the result back to Cloudflare or anywhere else.

  • Enterprise-grade and scalable

    Built for teams, with role-based permissions and dedicated support for Pro users. For details on Gumloop’s security practices, see trust.gumloop.com.

  • Start with a free trial

    You can test the Cloudflare MCP integration during Gumloop’s 14-day free trial before committing. Paid plans start at $37/month.

Frequently asked questions

Ship Cloudflare agents in minutes

Connect any AI agent to 100+ MCP servers, zero setup.
Book a demo
Gradient