What problem does it solve? Managing AI coding agent skills across platforms like Cursor, Claude Code, and Codex requires manually copying files into each tool's directory, and switching between separate CLIs and AI chat interfaces fragments the terminal workflow. ## Core Features & Use Cases - Skill Management: Install, update, and uninstall skills from GitHub repositories, then sync them to 51+ platform adapters including Cursor, Claude Code, Codex, TRAE, Cline, and Continue. - External CLI Management: Install and execute third-party command-line tools like gh and docker through the advai proxy via OpenCLI integration. - Terminal AI Chat: Run an interactive TUI chat session against any OpenAI-compatible API with configurable model, system prompt, agent, and in-session commands like /save and /model. - Knowledge Bases: Create local document collections, add files, and search them for reference during development. - Use Case: A team lead publishes a skills repository on GitHub; each engineer runs advai skill install with the repo URL and syncs the shared coding-standards skill to Cursor and Claude Code in one step. ## Quick Start Install advai-cli with pip, set your ADVAI_API_KEY environment variable, then ask the agent to install a skill from a GitHub repo and sync it to Cursor.