Skip to main content
Whop’s docs and API are built to be read by AI agents as well as people. Which surface you want depends on whether the agent needs to read Whop or act on it.

Pick a surface

MCP server

The Whop MCP server at mcp.whop.com/mcp operates on live Whop data and searches and reads this documentation. It acts on production only and can’t reach the sandbox, so test an integration by calling the sandbox API directly. It connects through a browser sign-in, so you never put an API key in the client config. One connection covers both, so there is no separate docs server to install. AI and MCP has the connection details and per-client setup.

Plugins

The Whop plugin bundles the Whop MCP server with skills that teach the agent how to use it.
Claude Code and Grok cover setup.

The skill

Whop’s docs publish an installable agent skill at docs.whop.com/skill.md. It teaches an agent what Whop is, when to reach for it, and where the API and SDKs live. Install it with the skills CLI:
Agents that support the skills convention discover it on their own through docs.whop.com/.well-known/agent-skills/index.json. The Whop CLI also ships its own skill for driving the CLI, installed with whop skills add and covered in Agent Mode.

Plain files for agents

You don’t need an MCP client to give an agent good context.
  • llms.txt indexes every page the way the docs navigation groups them, with a one-line description each. Large sections link to their own index files. Its agent instructions point to the MCP server, the CLI, and the Current API.
  • llms-full.txt is the same corpus expanded, for agents that want everything in one fetch.
  • Append .md to any docs URL to get the raw Markdown of that page. docs.whop.com/developer/quickstart.md returns the source of the quickstart, which is what you want in a context window instead of scraped HTML.

The CLI as an agent tool

The Whop CLI emits machine-readable output on every command, so an agent can call it directly and parse the result. That covers what the API doesn’t: scaffolding a project, running it locally, and deploying. Agent Mode covers the flags. Build with the CLI and Claude walks through it end to end.

Where to go next

Connect an MCP server

Setup for Claude, Cursor, and other MCP clients.

Build with the CLI and Claude

Scaffold and ship an app end to end.

Agent Mode

Skills sync and the machine-readable output flags.

Quickstart

The first API call, for you, or for your agent.