MCP AGENT DEMO · LIVE INVOCATION

Watch an Agent Use AINumbers

A real MCP session: the agent discovers the suite via .well-known/mcp.json, selects a tool, constructs a typed invocation, and receives a deterministic JSON response. Zero server calls. Zero inference. Client-side replay.

MCP v1+ AP2 v1.0 Deterministic Zero PII
Supersession notice (added 2026-07-27): On 17 April 2026 the Federal Reserve issued SR 26-2, Revised Guidance on Model Risk Management, which supersedes and replaces SR 11-7. Any reference to SR 11-7 on this page describes the framework as it stood before that date.
agent@ainumbers ~ mcp-session

What just happened

Five discrete steps: no magic, no inference. Here is exactly what each phase does.

01
Discovery
The agent reads .well-known/mcp.json at the root. It points to mcp/server.json for server capabilities and mcp/catalog.json for the full tool list with typed schemas. No API key needed: both are public and machine-readable.
.well-known/mcp.json
02
Typed Invocation
Every tool ships a manifest at manifests/{tool}.manifest.json, and each catalog entry carries the same inputSchema inline (JSON Schema Draft-07). The agent constructs a validated call: no prompt engineering, no natural-language parsing required.
JSON Schema · Draft-07
03
Deterministic Response
Each tool runs synchronously in a JS sandbox. Same inputs always produce the same outputs. No model inference, no hallucination risk. Deterministic and non-statistical, so it falls outside SR 26-2's definition of "model"; no model-risk validation regime applies. Results are AP2-envelope ready.
Deterministic · AP2 v1.0

Try it yourself

These tools appear in real agent invocation demos. Open any one and interact directly: client-side, zero PII.

MCP integration

Point your MCP-compatible agent at the AINumbers server endpoint. The registry loads once; tool schemas resolve on demand.

Connect once. Discover the tool set via .well-known/mcp.json, which resolves to mcp/catalog.json. Each tool's manifest under manifests/ provides the full inputSchema, outputSchema, and policy_mandate envelope spec.

Connect command
mcp connect https://ainumbers.co/mcp
.well-known/mcp.json · excerpt
{
  "schema_version": "well-known-mcp-v1",
  "servers": [
    {
      "id": "ainumbers-fintech-suite",
      "server_url": "https://ainumbers.co/mcp/server.json",
      "catalog_url": "https://ainumbers.co/mcp/catalog.json",
      "execution_model": "browser-reference"
    },
    {
      "id": "ainumbers-apps",
      "endpoint_url": "https://mcp.ainumbers.co/mcp",
      "execution_model": "mcp-apps-widgets"
    }
  ],
  "llms_txt": "https://ainumbers.co/llms.txt",
  "sitemap": "https://ainumbers.co/sitemap.xml"
}
mcp/catalog.json · one tool entry, abridged
{
  "name": "simulate_fraud_score",
  "inputSchema": {
    "type": "object",
    "required": ["amount"]
  },
  "metadata": {
    "tool_id": "fraud-score-simulator",
    "category": "fraud-risk",
    "url": "https://ainumbers.co/tools/04-fraud-score-simulator.html",
    "ap2_export": true,
    "execution_type": "browser-reference"
  }
}