Agent Clan

Overview

Agent Clan is a desktop workspace for working with AI agents from your own Claude Code or Codex accounts. It lets users select agents with gestures, speak tasks by voice, drag files or text onto agents as prompts, and monitor multiple agents across apps.

Agent Clan is an independent desktop workspace focused on working with AI agents through a visual, always-on-top interface. The site says users can connect their own Claude Code and Codex tools on their computer, then interact with agents as on-screen characters that stay visible over other apps. The product’s main workflow is choosing an agent, giving it a task by voice, and seeing the result without switching apps. The site describes gesture controls for bringing up agents, selecting one, and opening groups led by a leader agent. It also says mouse and keyboard can be used instead of gestures. Agent Clan presents agents as 3D characters in a room and supports multiple terminals, with each terminal representing another working agent. Users can customize each character’s appearance and organize agents into groups. The interface shows whether an agent is working, done, or waiting for approval, and users can allow or deny requests directly from the on-screen agent view. The site also describes drag-and-drop prompting: users can drop a picture, file, or text onto an agent to use it as the prompt. For voice and gesture input, the site says gesture recognition and speech-to-text run on the user’s computer, while transcribed text is sent to the chosen agent service. Pricing on the site includes a Free tier with 1 project and up to 2 agents, and a Pro plan described as a one-time purchase that adds unlimited projects and agents. The site states that AI service fees are separate and that users must use their own Claude Code or Codex account. It also notes that Agent Clan is independent and not affiliated with Anthropic or OpenAI.

Features

• Connects to the user’s own Claude Code and Codex accounts
• Always-on-top agent heads visible over any app
• Gesture-based agent selection and control
• Voice input for assigning tasks to agents
• Drag-and-drop prompts using files, images, or text
• Agent grouping with leader and crew structure
• 3D character and workspace visualization
• Status indicators for working, done, or waiting states
• Allow or deny agent requests without switching apps
• Pro supports unlimited projects and agents

Tags

FAQ