Software & website development
Back to list

Liuba's Agent AI - @liubas_helper_bot

Liuba's Agent AI - @liubas_helper_bot

📋 Project Status

Complete. Further improvements and integrations are possible.

Liuba's Agent AI — a lightweight Telegram assistant bot with basic commands and examples of integrating external services via free APIs. Created as a test assignment, but fully functional and currently NOT available for use: https://t.me/liubas_helper_bot . The bot provides quick access to useful information: weather, Wikipedia, GitHub, and even local semantic search in a knowledge base.

Entry point — src/main.py; bot logic is implemented in src/telegram_bot.py, and external modules are located in src/mcp/.


🧩 Main Features

  • /help — show brief help
  • /health — check system and API status
  • /weather <city> — current temperature (Open-Meteo API)
  • /wiki <topic> — brief Wikipedia summary
  • /github <query> — search GitHub repositories
  • /search <query> — semantic search in local knowledge base
  • /chat <text> — free dialogue with AI (OpenRouter)
  • /context 10 — show last 10 messages in chat
  • /forget — clear user history

🪄 MCP Integrations

ModuleDescriptionAPI
WeatherGet current temperature and coordinates by cityOpen-Meteo API
WikipediaBrief summaries of topicsWikipedia REST API
GitHubSearch public repositoriesGitHub REST API v3
Knowledge BaseSemantic search in local fact databaseChromaDB + SentenceTransformers
ChatFree dialogue with AIOpenRouter API

🛠️ Technologies Used

Language: Python 3.10+
Framework: python-telegram-bot (v20+)
Integrations: Open-Meteo, Wikipedia, GitHub, OpenRouter
Knowledge Base: ChromaDB + SentenceTransformers
Environment: dotenv

📦 Architecture

  • src/main.py — entry point
  • src/telegram_bot.py — command registration and polling
  • src/mcp/ — integration modules (weather, wiki, github, etc.)
  • src/knowledge_base.py — semantic search logic
  • Memory stored in memory.json (commands: /context, /forget)

🚀 Installation & Run

git clone <repo-url>
cd agent-liuba
python -m venv .venv
. .venv/Scripts/activate  # Windows PowerShell: . .venv/Scripts/Activate.ps1
pip install -r requirements.txt
python src/main.py

⚙️ Environment Variables

TELEGRAM_TOKEN=<your_bot_token>
WIKI_USER_AGENT=PomoshnikLiubyBot/1.0 (docs-example)
OPENROUTER_API_KEY=<your_openrouter_api_key>
OPENROUTER_MODEL=openrouter/auto

🐳 Docker

# Build
docker build -t agent-liuba .