Skip to content

devopsier/weather-assistant

Repository files navigation

Agentic Weather Assistant

AI weather assistant with intelligent tool-calling architecture.

Weather Assistant Demo

Features

  • Agentic AI: LLM decides which tools to use
  • Tool Calling: Weather and Air Quality APIs
  • Hallucination Management: Confidence scoring and risk assessment

Quick Start

  1. Setup:
cp .env.example .env
# Add your API keys to .env file
  1. Run:
go run . chat          # Interactive mode
go run . query "weather in Tokyo"  # Single query

API Keys Required

  • GEMINI_API_KEY - Google Gemini AI
  • WEATHERAPI_KEY - WeatherAPI.com (free 1M calls/month)

Architecture

main.go                      - Core logic, agentic LLM client
tools.go                     - Tool definitions and execution
cli.go                       - Command-line interface
weather_client.go            - Weather and air quality API clients
llm_hallucination_detector.go - AI risk assessment and confidence scoring

Example

You: What's the weather in Tokyo?
AI: I'll check the current weather data for you.
🌤️ Tokyo is currently 18°C with clear skies and good air quality (AQI: 42)

Built with Go, demonstrating agentic AI patterns, tool calling, and hallucination detection.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published