frankki-mcp
llms.txt1 starsChanged 2 days agoMITView on GitHub
FrankKi MCP is the official Model Context Protocol server for FrankKi (https://frankki.app): physical mail for AI agents. With one tool call, an agent composes, prices, sends, and tracks a real, printed letter to a physical address anywhere in the world. Letters are printed and delivered by our postal partners in each destination country. Hosted server, nothing to self-host. Key facts: - Endpoint: https://mcp.frankki.app (MCP streamable HTTP) - Registry name: app.frankki/letters (published in the Official MCP Registry, version 1.0.0) - Auth:…
Files
# FrankKi MCP > FrankKi MCP is the official Model Context Protocol server for FrankKi (https://frankki.app): physical mail for AI agents. With one tool call, an agent composes, prices, sends, and tracks a real, printed letter to a physical address anywhere in the world. Letters are printed and delivered by our postal partners in each destination country. Hosted server, nothing to self-host. Key facts: - Endpoint: https://mcp.frankki.app (MCP streamable HTTP) - Registry name: app.frankki/letters (published in the Official MCP Registry, version 1.0.0) - Auth: OAuth 2.1 with dynamic client registration, or API key as bearer token - OAuth metadata: https://mcp.frankki.app/.well-known/oauth-authorization-server - Billing: prepaid wallet (Stripe top-up); a German standard letter costs 1,99 EUR below 20 letters a month down to 1,59 EUR from 1,000 (volume ladder), top-ups 10 to 500 EUR, no VAT shown (§ 19 UStG), international priced per destination; call shipping_quote for live prices - Safety: per-key scopes and spending limits, maxCostEuros guard on sends, four-eyes human approval queue, sandbox mode with no real mail - Compliance: GoBD archiving, DATEV/DMS export, GDPR (EU data processing), German and English - Composition: FrankKi renders DIN 5008 letters from text; attachments supported; a finished letter PDF can be sent as it is via letter_upload_pdf (address-position check) then order_send with pdfStorageKey - Audience: B2B and developers (agents, backends, automated workflows); consumer app available on iOS ## Docs - [README](https://github.com/datPascal/frankki-mcp/blob/main/README.md): overview, quick starts, FAQ - [Tool reference](https://github.com/datPascal/frankki-mcp/blob/main/docs/tools.md): all tools with scopes and side effects - [Authentication](https://github.com/datPascal/frankki-mcp/blob/main/docs/authentication.md): OAuth 2.1, API keys, scope list - [Client setup](https://github.com/datPascal/frankki-mcp/blob/main/docs/clients.md): Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, OpenAI Agents SDK, LangChain - [Use cases](https://github.com/datPascal/frankki-mcp/blob/main/docs/use-cases.md): dunning, terminations, compliance mail, onboarding ## Links - [FrankKi homepage](https://frankki.app) - [MCP for humans and agents](https://frankki.app/mcp) ([English](https://frankki.app/en/mcp)) - [Site llms.txt](https://frankki.app/llms.txt)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.