How lehakrechet/tg_bot_ai works
Project type and purpose
• This is a Python‐based backend service—a Telegram AI chatbot (“tg_bot_ai”).
• It integrates with Telegram’s Bot API, a third‐party AI (YandexGPT), and an internal “deepseek” (dsk) library that provides cryptographic or search capabilities via WebAssembly and Python modules.
File structure
• Top level:
• deepseek.py – likely a helper wrapper around the dsk package
• tg_bot.py – main entry point for the Telegram bot
• yandexgpt.py – adapter for Yandex GPT API
• file_worker.py – handles file I/O or background file tasks
• test.py/test.txt – ad hoc test harnesses
• recomendation.txt, requirements.txt
• dsk/ directory:
• api.py – core API calls for deepseek functionality
• pow.py – proof‐of‐work or crypto submodule
• wasm/sha3_wasm_bg…wasm – WebAssembly binary for SHA3
• history/ directory: persists session transcripts
README
• Only title provided (“tg_bot_ai”).
• No in‐depth architecture notes; inference must come from code structure.