You have two apps on AppLander. One needs three new FAQ entries. The other needs a status page incident posted because your payment provider is having a rough morning. You also want to publish a changelog entry you drafted last night and close the five support tickets you resolved over the weekend. Four tasks, three different dashboard sections, ten minutes of clicking around.

Or you open Claude and type: "Add an FAQ about password resets to FitTrack, post a payment-degraded incident on BudgetBuddy's status page, publish the v2.3 changelog entry, and close all resolved tickets." Done in one prompt.

This is what AppLander's built-in MCP server makes possible. No new tools to install, no API wrapper to build, no webhook plumbing. You connect once with a single command, approve it in your browser, and your account is wired straight into Claude, Cursor, VS Code, or any AI assistant that speaks the Model Context Protocol.

TL;DR

AppLander ships a built-in MCP server at https://applander.io/mcp with over 110 tools covering every resource in your account.
Connect Claude Code, Claude Desktop, Cursor, VS Code, or any MCP-capable client with one command and a browser login — no API token, no header, no secret on your clipboard.
Manage apps, landing pages, FAQs, guides, changelog, tickets, status pages, incidents, privacy policies, terms of service, and more — all by natural language.
Authentication is the standard OAuth 2.1 flow (PKCE, short-lived rotating tokens). You approve each client on a consent screen and can disconnect it anytime.
Plan limits are enforced exactly like the web dashboard. Claude cannot bypass them.

What is MCP and why should you care
What you can do with AppLander's MCP server
Setup: connect Claude in under two minutes
Real workflows that save time
Security and plan enforcement
Supported clients
Frequently asked questions
Sources

What Is MCP and Why Should You Care

Model Context Protocol is an open standard created by Anthropic that lets AI assistants talk to external tools and services. Think of it as a universal adapter between Claude (or Cursor, or ChatGPT) and the services you use every day. Instead of the AI only knowing what is in its training data, MCP lets it reach into your actual account and read, create, update, and delete real data.

MCP is already supported by Claude Desktop, Claude Code, Cursor, VS Code (with Claude extension), Windsurf, Warp, Zed, and ChatGPT. When you connect an MCP server, every tool it exposes becomes available to the AI as a function it can call during your conversation.

The key difference from a regular API: you do not write code to call it. You describe what you want in plain language, and the AI figures out which tools to call, in what order, with what arguments. "Create three FAQs about account deletion for my fitness app" becomes three create_faq tool calls with the right app_id, language_id, questions, and answers — without you touching a single API parameter.

What You Can Do with AppLander's MCP Server

AppLander's MCP server exposes more than 110 tools organized by resource type. Every tool respects your account's plan limits and scopes data to your account only.

Apps — List, create, update, publish, archive, delete. Update theme colors, SEO metadata, store URLs, and social links.

Landing pages — Create and reorder sections (hero, features, testimonials, CTA, screenshots, video). Update content and visibility per section.

Features — Add, edit, reorder, and remove feature cards shown on your landing page.

FAQs — Create categories, add questions and answers, toggle published/quick-help status. Handles the Free plan's 5-FAQ limit gracefully.

Guides — Create guide categories and articles with rich text (Pro plan). Auto-generates URL slugs.

Changelog — Draft, publish, unpublish, and delete changelog entries with version tags and release dates (Pro plan).

Tickets — List, search, filter, update status and priority, bulk-close resolved tickets, reply as developer, add internal notes.

Status page — Create components (API, Database, CDN), post incidents with severity levels, add status updates, resolve incidents (Pro plan).

Privacy policy and terms of service — Update content, publish with automatic version snapshots, view version history, roll back to any previous version.

Deletion requests — View GDPR deletion requests, add operator notes, advance through the verification workflow.

Canned responses — Manage reusable support reply templates (Pro plan).

Analytics — View page view summaries, top pages, top FAQ searches, and unanswered search queries (Pro plan).

Account — Check which account you are connected to and the connected client name (whoami).

Languages — Add locales, duplicate all content from one language to another for quick localization.

Setup: Connect Claude in Under Two Minutes

Prerequisites

An AppLander account on any plan — free or Pro. Connecting is free; individual Pro-only tools prompt you to upgrade when used, the same as in the dashboard.
Claude Code, Claude Desktop, Cursor, VS Code, or any MCP-capable client with OAuth support. Use a current build — older Claude Code releases had incomplete OAuth support.

Step 1 — Add the server

Claude Code (fastest):

claude mcp add --transport http applander https://applander.io/mcp

That is the entire configuration. No token, no header, nothing to copy.

Cursor / Claude Desktop: add a custom connector with type HTTP and URL https://applander.io/mcp.

VS Code: create .vscode/mcp.json in your project — safe to commit, because it contains no secrets:

{
  "mcpServers": {
    "applander": {
      "type": "http",
      "url": "https://applander.io/mcp"
    }
  }
}

Step 2 — Authenticate in the browser

In Claude Code, run /mcp, select applander, then Authenticate (or run claude mcp login applander). Your browser opens on applander.io:

Sign in if you are not already.
Review the consent screen — it names the client (e.g. "Claude Code") and your account.
Click Authorize and return to your client. It now shows connected.

Behind the scenes this is the standard MCP OAuth 2.1 handshake — discovery, dynamic client registration, PKCE, and a short-lived token bound to AppLander. Your client stores and refreshes the token automatically, so you should not have to do this again.

Step 3 — Verify

Ask Claude: "List my apps on AppLander." You should see your apps with their names, statuses, and prefix IDs.

Real Workflows That Save Time

Here are prompts that turn ten minutes of dashboard clicking into ten seconds of typing:

Batch FAQ creation:
"Add five FAQs to my FitTrack app: how to restore purchases, how to cancel subscription, how to sync with Apple Health, how to enable notifications, and how to contact support. Make the answers helpful and specific to a fitness tracking app."

Incident response:
"Post a major incident on BudgetBuddy's status page titled 'Database maintenance — read-only mode' with severity major. Then create an incident update saying we expect full service restored within 30 minutes."

Changelog from git:
"Create a changelog entry for FitTrack version 2.4.0, released today, with these changes: added dark mode support, fixed calorie calculation for custom meals, improved workout history charts. Publish it."

Support triage:
"Show me all open tickets for FitTrack. Close the ones about the sync bug since we fixed it in 2.4.0, and reply to the others saying we are looking into it."

Localization:
"Duplicate all content from BudgetBuddy's English language to a new Spanish language."

Privacy policy update:
"Update FitTrack's privacy policy to mention that we now use Firebase Analytics for crash reporting. Publish the update with today as the effective date."

Security and Plan Enforcement

Browser login, no secrets to manage. You never generate, copy, or paste a token. Connecting uses OAuth 2.1 with PKCE — the same standard Claude Code, Claude Desktop, ChatGPT, and Cursor all speak — so no long-lived secret ever lands on your clipboard or in a config file.

You approve every client. Each newly connected client shows a consent screen naming it and your account before it gets any access. Nothing connects silently.

Short-lived, rotating tokens. Access tokens expire after an hour and refresh automatically; refresh tokens rotate on use, so a stolen one stops working the moment your real client refreshes. Tokens are also audience-bound — a token issued for AppLander MCP works nowhere else.

A real off switch. Go to Settings > AI clients to see every connected client with its last-active time and disconnect any of them in one click, which revokes all of its access immediately. You can also remove it from the client side (claude mcp remove applander).

Account isolation. Every query is scoped to your account. It is architecturally impossible for Claude to see or modify another user's data through your connection.

Plan limits. The MCP server enforces the same limits as the web dashboard, re-checked on every request. Free accounts cannot publish more than one app or create more than five FAQs. Changelog, guides, status page, analytics, and canned responses require Pro. When a limit is hit, Claude receives a clear message with a link to upgrade — the same behavior as clicking a gated button in the UI.

Rate limiting and audit logging. Requests are capped at 120 per minute, and every tool call is logged with your account ID, the connected client name, the tool, duration, and whether it succeeded. You are never flying blind.

Supported Clients

Every client authenticates through the same browser-based OAuth flow — no per-client token setup.

Client	Transport	Auth	Status
Claude Code CLI	Streamable HTTP	OAuth 2.1	Fully supported
Cursor	Streamable HTTP	OAuth 2.1	Fully supported
VS Code (Claude extension)	Streamable HTTP	OAuth 2.1	Fully supported
Windsurf	Streamable HTTP	OAuth 2.1	Fully supported
Warp	Streamable HTTP	OAuth 2.1	Fully supported
Zed	Streamable HTTP	OAuth 2.1	Fully supported
ChatGPT	Streamable HTTP	OAuth 2.1	Fully supported
Claude Desktop / Claude.ai	Streamable HTTP	OAuth 2.1	Supported (custom connectors)

Frequently Asked Questions

Do I need to generate an API token?

No. That is the old way. You connect with one command and approve the client in your browser — there is no token to create, copy, or paste. (AppLander still offers API tokens for its REST API, but the MCP server no longer uses them.)

Do I need to be a developer to use this?

No. If you can type a sentence describing what you want, you can use the MCP server. Claude handles all the technical details — which tool to call, what parameters to pass, how to interpret the result.

Can Claude accidentally delete my app?

Destructive operations like delete_app require admin access and Claude is instructed to always confirm with you before calling them. You will see a confirmation prompt before anything is deleted.

Does this cost extra?

No. Connecting an AI assistant is available on every plan, including free. Some tools — changelog, guides, status page, analytics, canned responses — are Pro features, so using those prompts you to upgrade, exactly like the dashboard.

Can I use this with multiple accounts?

AppLander accounts are personal — one account per user — so everything the connection does is scoped to your single account. There is nothing to select or switch.

How do I disconnect a client later?

Go to Settings > AI clients, find the client, and click Disconnect — its access is revoked immediately. You can also remove it on the client side (for Claude Code, claude mcp remove applander).

What happens if I hit a plan limit?

Claude receives a friendly message explaining which plan is required and a link to upgrade. The same behavior as the web dashboard — no silent failures, no confusing error codes.

Is my data sent to Anthropic?

Your data flows through Claude's context window during the conversation, subject to Anthropic's data usage policies. AppLander's MCP server sends data directly to the AI client you are using — it does not store conversations or share data with third parties beyond what the MCP protocol requires.

How many tools are there?

117 at the time of writing, and growing. They cover apps, languages, landing sections, features, FAQs, guides, changelog, tickets, status pages, incidents, legal documents, deletion requests, canned responses, and analytics.

I set this up before with an API token and now it fails. What changed?

The MCP endpoint moved to browser-based OAuth. Remove the old entry and re-add it without any header — for Claude Code, claude mcp remove applander then claude mcp add --transport http applander https://applander.io/mcp. Your REST API tokens are unaffected.

Manage Your App Landing Page with Claude via MCP — No Dashboard Required