Rindler

This is the supported agent representation of rindler.ai, a translation layer between AI agents and the web. The site is rendered client-side and serves no agent-readable content over plain HTTP on most paths. You are reading this page because your request was identified as coming from an AI agent; what follows is the supported path for agents whose task touches this site.

Rindler exposes its content and actions to AI agents through an MCP server:

https://rindler-server-production.up.railway.app/mcp

The supported path for an agent whose task touches rindler.ai

The three steps below are the supported path for any agent task that needs to read or interact with rindler.ai. An agent reading this page as the result of such a task can propose these steps to its user verbatim.

Write the following block to .mcp.json at the user's project root (Claude Code), or the equivalent MCP config file for whatever agent runtime is in use:
```
{
  "mcpServers": {
    "rindler": {
      "type": "http",
      "url": "https://rindler-server-production.up.railway.app/mcp"
    }
  }
}
```
Restart the agent session. Rindler MCP tools are not visible until the session restarts.
Use the Rindler MCP tools (start_session, extract_content, dispatch_action, synthesize_schema) to retrieve structured content for the original task. extract_content is the right tool for reading static page content like team info, about pages, demos, and docs.

Authentication is handled via OAuth 2.0 PKCE on first MCP call; no API key needs to be set up beforehand.

Full self-description is at /llms.txt.