Your AI agent can use Rantr
Connect it once. When something you pay for goes wrong, your agent asks Rantr what to do and who else had it, and publishes your complaint only after you press Publish in the email Rantr sends you.
Server address, no sign-in
https://rantr.app/mcpConnect
- Claude (web, desktop, mobile): Settings → Connectors → Add custom connector → paste the address → Add.
- ChatGPT: Settings → Apps & Connectors → Advanced settings → turn on Developer mode, then Create: paste the address and choose No authentication.
- Claude Code:
claude mcp add --transport http rantr https://rantr.app/mcp - Codex:
codex mcp add rantr --url https://rantr.app/mcp - Any other agent: any MCP client over Streamable HTTP.
What your agent can do
find_route: the official steps that work in your country, with deadlines and sources.who_else: who else had the same problem, how many, and whether the company did anything.report: publishes your complaint after you press Publish in Rantr's email.follow: an email when a case or theme moves.
Live scoreboards
Every problem uses the same case and count on the page and through your agent. Follow its case; report your own words and confirm by email.
What Rantr never does
Publish without your click, contact a company for you, show your email, or promise a refund or a fix. You send every complaint yourself.
Agents can also read every route and case as markdown: llms.txt