← MoneyChoice

Your agent fetches the Agent Card, sends a JSON-RPC request (macro forecast, scenario, or risk), and gets back a completed task or streamed response.

Agent-to-Agent

Macro, scenarios & risk over A2A

One endpoint for macro indicators, scenario analysis, and risk assessment. Your agents discover capabilities via the Agent Card and send tasks to get results or streams.

What’s in it for you

Fed-grade data. CPI, unemployment, rates, GDP—same sources as the Fed, no secondhand feeds.
One API, three uses. Macro forecasts, scenario analysis, and risk assessment from a single endpoint.
Use anywhere. Your AI assistant, dashboard, script, or app—request in plain language or structured data.
Backtested, not guesswork. We pick the best model (monthly/quarterly/yearly) by historical error.
Streaming. Get results as they’re ready with message/stream for long-running tasks.

How people use it

Ask your AI assistant
You ask: “What’s the inflation forecast for the next 6 months?” Your agent calls our endpoint with the right request.
You get: CPI forecast with dates and values, or a short summary—in your chat or app.
Power a dashboard or report
Your app requests forecasts for unemployment, 10-year yield, or GDP. You display them in charts or tables.
You get: Structured forecast data (periods and values) to plot or export.
Get a risk assessment
You send: “Risk assessment for inflation and rates.” The agent routes to our risk skill.
You get: A written risk analysis you can show clients or use in decisions.
Automate research with code
Your script or pipeline POSTs a JSON-RPC request (e.g. inflation_cpi, monthly). No UI needed.
You get: JSON with forecast data—same quality as the Fed’s sources, one call.

How to use it

1 Get Agent Card from .well-known/agent.json
2 POST JSON-RPC to endpoint (message/send or message/stream)
3 Read task result or stream in artifacts

How it works

Fetch Agent Card → Send JSON-RPC → We run task → Result or stream
Same data as the Fed JSON-RPC 2.0 Streaming

Three skills, one endpoint

Macro indicators

Forecasts for CPI, unemployment, rates, GDP. Request by indicator name and time horizon (monthly/quarterly/yearly).

Scenarios

Scenario analysis for your use case. Send a text prompt; the agent routes to the right skill.

Risk assessment

Risk assessment for inflation, rates, and macro factors. Same endpoint, natural language.

Why use this agent

  • Same data as the Fed. Official macro indicators—CPI, unemployment, rates, GDP—no secondhand sources.
  • Standards-based. A2A Agent Card and JSON-RPC 2.0; works with any A2A client.
  • Streaming. Use message/stream with Accept: text/event-stream for SSE results.
  • Three skills. Macro indicators (forecasts), scenarios (analysis), risk assessment—one agent, one endpoint.

Discovery (Agent Card)

RFC 8615 well-known path. Use this URL to discover capabilities and skills.

Agent Card URL
https://api.moneychoice.us/.well-known/agent.json

A2A endpoint (JSON-RPC 2.0)

POST JSON-RPC body. Methods: message/send, message/stream, tasks/get.

Endpoint
POST https://api.moneychoice.us/moneychoice_complete.api?x=a2a

API docs (A2A) →

Examples

Macro forecast (message/send)

Request CPI forecast; get a completed task with forecast data in the artifact.

You send
{"method":"message/send", "params":{"message":{"parts":[{"kind":"data","data":{"indicator_name":"inflation_cpi","time_horizon":"monthly"}}]}}}
→
You get
Task with status.state: "completed" and artifacts[].parts with forecast periods & values (kind: "data")
Request (POST body) 
{
  "jsonrpc": "2.0",
  "id": "1",
  "method": "message/send",
  "params": {
    "message": {
      "role": "user",
      "messageId": "msg-1",
      "contextId": "my-context-123",
      "parts": [
        { "kind": "text", "text": "Get inflation forecast" },
        { "kind": "data", "data": { "indicator_name": "inflation_cpi", "time_horizon": "monthly" } }
      ]
    }
  }
}

Response: result is a task with status.state: "completed" and artifacts[].parts containing kind: "data" (forecast periods/values) and kind: "text".

Scenario or risk (text prompt)

Send a text part; the agent routes to the right skill.

Request (POST body) 
{
  "jsonrpc": "2.0",
  "id": "2",
  "method": "message/send",
  "params": {
    "message": {
      "role": "user",
      "messageId": "msg-2",
      "contextId": "my-context-123",
      "parts": [ { "kind": "text", "text": "Risk assessment for inflation and rates" } ]
    }
  }
}

Streaming (message/stream)

Same params as above; set method: "message/stream" and header Accept: text/event-stream. You get SSE lines: status-update, then task result, then data: [DONE].

curl (macro forecast) 
curl -X POST "https://api.moneychoice.us/moneychoice_complete.api?x=a2a" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":"1","method":"message/send","params":{"message":{"role":"user","messageId":"m1","contextId":"ctx1","parts":[{"kind":"data","data":{"indicator_name":"inflation_cpi","time_horizon":"monthly"}}]}}}'

MCP (macro forecasts only)

For MCP-only macro economic forecasts in your AI (Claude, Cursor, etc.), use the MCP endpoint.

MCP page

https://api.moneychoice.us/mcp — Claude, Cursor