2 min read
Send a message
curl -s -X POST https://api.credicorp.co.uk/public/v1/support/chat \
-H 'Content-Type: application/json' \
-d '{
"conversation_id": "conv_8a1c",
"message": "What is the maximum I can borrow for six months?"
}'The response carries the assistant reply and the conversation_id to reuse on the next turn. Keep sending the same id to preserve context across the conversation.
Wire it into your UI
async function ask(conversationId, message) {
const res = await fetch(`${BASE}/support/chat`, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ conversation_id: conversationId, message }),
});
const { reply } = await res.json();
return reply;
}
Frequently asked questions
How do I keep conversation context?
Reuse the same conversation_id across turns. Start a new conversation by omitting it on the first message and storing the id the response returns.
Is this the same as the MCP server?
No. This is a support chat surface for end users. The MCP server exposes tools to AI agents. They serve different audiences.
Related reading

Quickstart: fetch the support widget configuration
GET /public/v1/support/widget returns the configuration for the embeddable Credicorp support widget. Read the…
Read →
Quickstart: connect an AI agent to the Credicorp MCP server
POST /public/v1/mcp is a Model Context Protocol endpoint that exposes the public API as agent tools. Point an…
Read →
Quickstart: fetch published CMS content by key
GET /public/v1/cms/pages/{key} serves published Credicorp content by a stable key. Use it to pull canonical…
Read →Funding for UK limited companies
Credicorp lends to your company, not to you personally — short-term working capital with no personal guarantee. See what your business could access.