Enables voice conversations and screen capture for AI agents via MCP-compatible clients, using Pipecat for speech-to-text and text-to-speech, with support for browser, WebRTC, and phone transport.
Enables MCP clients to drive ChatGPT's web interface through a built-in persistent browser, requiring only a single manual login. Prompts are sent through temporary incognito chats and the captured answer text is returned as tool results, along with live model listing, login/status probing, and optional HTTP transport.