Ern spricht jetzt mit Ihrer KI

A new interface — not for humans               

                                                                                                           
Ern has a new kind of interface. The Ern MCP server connects any AI assistant that supports the Mode Context Protocol to live protocol data. Ask about current APYs, vault routing, or simulate years of DCA — in natural language. No API key. No authentication. Setup takes 30 seconds.

What is MCP                                                                                   

The Model Context Protocol is an open standard that lets AI tools call external services directly. 
Instead of copy-pasting numbers from a dashboard, you ask your AI assistant a question and it pulls the answer straight from the protocol.

Ern's MCP server exposes four tools:                                                                       

- Protocol overview — TVL, fee structure, live supply APY, and a summary of every active vault.          

- Vault list — All active vaults with addresses, TVL, underlying assets, and current provider allocations.

- Vault details — Full detail for a single vault, including allocation history and back-tested APY across 1 to 10 year windows.                                                                                   

- Growth curve — DCA growth simulation for any deposit amount, lookback period, and granularity.

Every response comes from live on-chain data. The same data that powers ern.app.

What you can ask                               

Once connected, you talk to your AI the way you normally would. The MCP handles the rest.                 

• "What's the live APY on USDC?"                                                                         
• "Simulate $10k DCA'd over 5 years."   
• "Which provider is each vault routed to?"
• "Compare the 3-year and 5-year back-tested returns on the USDC vault."
• "How much BTC would I have accumulated on a $50k deposit over 10 years?"                               

The AI fetches the data, does the math, and returns a plain-language answer. No dashboards. No spreadsheets. Just a question and an answer.              

Works with everything                          

The server is compatible with any client that supports MCP:

- Claude Code — one command: claude mcp add ern-protocol https://use.ern.app/mcp/sse                     

- Claude Desktop — add the server URL in Settings > Developer > Edit Config.

- Cursor — Settings > MCP > Add new global MCP server.                                                   

- VS Code — run MCP: Open User Configuration or add to .vscode/mcp.json.

- Windsurf — click the hammer icon in Cascade > Configure.                                               

Public REST API                                

The same data is also available as a standard REST API for developers building their own integrations.

- GET /protocol — protocol-level snapshot: TVL, fees, live APY, vault summary.                           

- GET /vaults — all active vaults with current on-chain state.

- GET /vaults/{slug} — full vault detail with allocation history and trailing APY.                       

- GET /vaults/{slug}/growth_curve — DCA growth simulation with customizable parameters.                  

Base URL: https://use.ern.app/api/public/v1                                                              

No authentication required. Rate limits are 60 requests per minute, 500 per hour, 5,000 per day. MCP requests are limited to 30 per minute and 300 per hour.   

Why this matters                                                                              

DeFi protocols generate a lot of data. Most of it sits behind dashboards that require context to interpret. MCP removes that friction. The protocol's data becomes a conversation — accessible to anyone with an AI assistant and a question.

This is the same principle behind Ern itself. Stablecoins already earn yield. Ern routes that yield into Bitcoin. Now, understanding what that means for your position is as simple as asking.

Full docs and install instructions: use.ern.app/api/public/v1/docs