AI and Agents

CPFHub.io was built to work natively with AI agents. It has official support for the main frameworks and protocols used by modern LLMs. A CPF is Brazil's individual taxpayer ID, and Receita Federal is Brazil's federal tax authority.

What is available

Tool	Description	Installation
MCP Server	Official Model Context Protocol server	`https://api.cpfhub.io/mcp`
Agent Skills	Tool definitions for Claude, GPT-4 and others	Agent Skills
OpenAPI 3.1	API spec for automatic client generation	`/openapi.json`
llms.txt	Machine-readable index of the documentation	`/llms.txt`

Use cases with AI agents

Conversational onboarding: An onboarding agent asks for the user's CPF during the chat and validates their identity on the spot, without leaving the conversation flow.

Autonomous fraud prevention: A compliance agent monitors transactions and automatically looks up suspicious identities before approving.

Customer support: A support assistant confirms the customer's identity by CPF before accessing sensitive account data.

No-code automations: n8n, Make or Zapier flows that validate CPFs as part of larger lead capture or signup workflows.

Available tools

The remote MCP Server exposes three tools for AI agents:

Tool	Description	Cost
`lookup_cpf`	Simple Lookup: name, gender and date of birth from a CPF	1 credit per CPF found
`lookup_cpf_realtime`	Real-Time Lookup at Receita Federal (CPF + date of birth): registration status, year of death (`deathYear`, an integer or `null`), control code and proof	1.5 credits per successful lookup
`get_quota_info`	Plan, remaining and used credits and billing status	Free

The batch lookup has no MCP tool: use POST /cpf/bulk through REST.

Next steps

MCP Server: set up the MCP server in Claude Desktop, Claude Code, Cursor or any MCP client
Agent Skills: tool definitions to use directly in system prompts
Prompt Library: ready-made prompts for common use cases
Platforms: n8n, Make, Zapier, LangChain

Updated on October 3, 2026