Pular para conteúdo

Cidadao.AI Backend

Multi-Agent AI System for Brazilian Government Transparency

Production Tests Coverage Agents

What is Cidadao.AI?

Cidadao.AI is a production multi-agent AI system designed to analyze Brazilian government transparency data. It features 17 specialized AI agents with unique Brazilian cultural identities, each responsible for different aspects of data investigation, analysis, and reporting.

Production Status

Live at: cidadao-api-production.up.railway.app

  • 99.9% uptime since October 2025
  • 17 agents operational (16 functional + 1 base framework)
  • 323+ API endpoints
  • PostgreSQL + Redis infrastructure

  • 🤖 Agents

    Explore our 17 AI agents with Brazilian cultural identities

    View Agents

  • 🏗 Architecture

    Understand the multi-agent system design

    Architecture

  • 🔌 API Reference

    323+ endpoints for transparency data

    API Docs

  • 🚀 Deployment

    Deploy to Railway, Docker, or cloud

    Deploy Guide

Agent Role Coverage
Zumbi dos Palmares Anomaly Detection 96.32%
Anita Garibaldi Pattern Analysis 94.87%
Oxossi Data Hunting 94.44%
Abaporu Master Orchestration 86%
Obaluaie Corruption Detection 81%

View all 17 agents

Getting Started

Installation

# Clone repository
git clone https://github.com/anderson-ufrj/cidadao.ai-backend.git
cd cidadao.ai-backend

# Install dependencies
make install-dev

# Configure environment
cp .env.example .env
# Edit .env with your API keys (MARITACA_API_KEY or ANTHROPIC_API_KEY)

Run Development Server

make run-dev
# API available at http://localhost:8000/docs

Run Tests

JWT_SECRET_KEY=test SECRET_KEY=test make test

Key Features

Multi-Agent System

  • 17 specialized agents with unique Brazilian cultural identities
  • Reflective architecture with quality thresholds and automatic improvement
  • Lazy loading for 367x faster startup (3.81ms vs 1460ms)

Government Data Integration

  • 30+ transparency APIs integrated (federal, state, municipal)
  • Portal da Transparencia integration
  • Real-time contract investigation

Production-Ready

  • PostgreSQL + Redis infrastructure
  • Prometheus + Grafana monitoring
  • Railway deployment with 99.9% uptime

Project Status

pie title Agent Coverage Distribution
    "Tier 1 (>90%)" : 10
    "Tier 2 (80-89%)" : 5
    "Tier 3 (70-79%)" : 1
    "Base Framework" : 1

Current Sprint

December 2025: Focus on test coverage improvement, additional government APIs, and frontend integration enhancements.

Documentation Structure

Section Description
Agents All 17 agents with implementation details
Architecture System design, patterns, and diagrams
API REST endpoints, streaming, WebSocket
Development Setup, contributing, best practices
Deployment Railway, Docker, production guides
Project Roadmap, status, planning

Contributing

We welcome contributions! See our Contributing Guide for details.

# Run checks before committing
make check

Documentation Update