The Model Context Protocol is an open standard for connecting AI assistants to tools and data. The Creditcorp MCP endpoint lets a compatible agent use the public API without custom glue code.

Quickstart: connect an AI agent to the

Q: How does the agent discover tools?

Through the standard MCP tools/list call, which returns each tool's name and JSON schema so the agent can call it correctly.

2 min read

POST/public/v1/mcp

MCPModel Context Protocol

toolsProducts, pricing, quote

What the server exposes

The MCP endpoint advertises a tool set mapped onto the public ring — for example list_products, get_pricing and quote_loan. An agent discovers them via the standard MCP handshake and calls them like any other tool.

Point a client at it

Any MCP client can connect. The transport is JSON-RPC over HTTP POST to the endpoint:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/list",
  "params": {}
}

The server responds with the available tools and their JSON schemas; the agent then issues tools/call requests to run them.

Keep it read-mostly

The public MCP surface is deliberately read-oriented — catalogue, pricing and quotes. Actions that create records or move money live behind the authenticated partner ring and are never exposed to an anonymous agent.

Frequently asked questions

What is MCP?

The Model Context Protocol is an open standard for connecting AI assistants to tools and data. The Credicorp MCP endpoint lets a compatible agent use the public API without custom glue code.

Can the agent submit an application?

No. The public MCP surface is read-mostly — catalogue, pricing and quotes. Anything that writes a record or moves money requires the authenticated partner ring.

How does the agent discover tools?

Through the standard MCP tools/list call, which returns each tool's name and JSON schema so the agent can call it correctly.