The Telepath team released Television, an open source visual workspace where coding agents like Claude Code and Codex can create and modify artifacts on a shared screen. A CLI/server runs alongside the agent harness while a macOS client (or browser) displays artifacts; any inference provider the agent already uses is reused. The repo opened to the public after testing with a few hundred alpha users.
Lody is an open-source, local-first shared workspace that connects coding agents like Claude Code and Codex across laptops, workstations, servers, and cloud VMs. It centralizes conversations, runtime status, files, diffs, and decisions, solving the problem of fragmented visibility and collaboration around agent work. It does not replace agents or force hosted execution, but provides a shared context layer for teams.
Lody is an open-source, local-first shared workspace that connects coding agents like Claude Code and Codex across laptops, workstations, servers, and cloud VMs. It centralizes conversations, runtime status, files, diffs, and decisions, solving the problem of fragmented visibility and collaboration around agent work. It does not replace agents or force hosted execution, but provides a shared context layer for teams.
TalkToMe is an open-source macOS app that lets an AI coding agent place a phone call to you from its current session, so you can talk with it and hear spoken replies while the agent keeps its tools, files, and history. It supports agents that run shell commands, such as Codex, Claude Code, and Hermes, and uses ElevenLabs or local models for the voice conversation. It solves the problem of staying hands-free while an agent needs your input during a long-running task.
TalkToMe is an open-source macOS app that lets an AI coding agent place a phone call to you from its current session, so you can talk with it and hear spoken replies while the agent keeps its tools, files, and history. It supports agents that run shell commands, such as Codex, Claude Code, and Hermes, and uses ElevenLabs or local models for the voice conversation. It solves the problem of staying hands-free while an agent needs your input during a long-running task.
WorkBraid is a local CLI/MCP tool that renders visual architecture diagrams and change proposals so humans can review agent-proposed code changes. Agents operate it via MCP to submit visual change proposals, which are reviewed by humans or AI and backed by git. It targets developers who struggle to evaluate large text-only diffs from coding agents.
WorkBraid is a local CLI/MCP tool that renders visual architecture diagrams and change proposals so humans can review agent-proposed code changes. Agents operate it via MCP to submit visual change proposals, which are reviewed by humans or AI and backed by git. It targets developers who struggle to evaluate large text-only diffs from coding agents.
Dax Shepherd is a macOS tool for orchestrating multiple coding agents (Claude Code, Codex) across projects, extending the Herdr concept. Agents' sessions stay running while developers browse and navigate the server filesystem through a GUI. It targets developers running herds of coding agents who need visibility and organization.
Dax Shepherd is a macOS tool for orchestrating multiple coding agents (Claude Code, Codex) across projects, extending the Herdr concept. Agents' sessions stay running while developers browse and navigate the server filesystem through a GUI. It targets developers running herds of coding agents who need visibility and organization.
Codewindow is a tool that provides a picture-in-picture view for terminal-based coding agents like Codex, allowing users to monitor agent activity in a floating, resizable window while working in other applications. It solves the problem of missing bugs or bad practices introduced by agents when not continuously watched, making it useful for developers who want to supervise long-running coding tasks without constant attention.
Codewindow is a tool that provides a picture-in-picture view for terminal-based coding agents like Codex, allowing users to monitor agent activity in a floating, resizable window while working in other applications. It solves the problem of missing bugs or bad practices introduced by agents when not continuously watched, making it useful for developers who want to supervise long-running coding tasks without constant attention.