nanobot-ai

Command-line interface and Python module entry point for nanobot, handling interactive terminal input, gateway server startup, and agent commands via Typer and prompt_toolkit

Configuration schema, loader, runtime path resolution, and workspace template files (AGENTS.md, SOUL.md, USER.md, TOOLS.md, HEARTBEAT.md, MEMORY.md) that bootstrap the agent's identity and behavior

Core agent engine including the agent loop, LLM prompt context builder, memory consolidation, skills loader, subagent manager, and all tool implementations (filesystem, shell, web, MCP, message, cron, spawn)

Messaging platform integrations (Telegram, Slack, Discord, DingTalk, Feishu, Email, Matrix, WeCom, WhatsApp, QQ, Mochat) and the async message bus that decouples channels from the agent core

LLM provider abstractions (LiteLLM, OpenAI Codex, Azure OpenAI, Custom), cron scheduling, heartbeat proactive tasks, session persistence, network security, provider registry, voice transcription, and shared utility functions

Bundled agent skills (GitHub, weather, summarize, tmux, ClawHub, memory, cron, skill-creator) with SKILL.md definitions and helper scripts, plus project documentation (README, CONTRIBUTING, SECURITY, onboarding guides) and build/deploy configuration

Comprehensive test suite covering all nanobot subsystems: channel integrations (Telegram, Slack, DingTalk, Feishu, Matrix, Discord, QQ, Email), agent loop, memory consolidation, tool validation, provider behavior, security, CLI commands, and configuration