Paradigm-Study/claude-share

Multiplayer sessions for Claude Code. Share a link, work together live, and review the same replies, tools and previews.

What it solves

Claude Code normally works alone on one machine. This project lets two or more people work in the same Claude session, seeing each other's prompts, streamed replies, and tool results, while only the host's computer performs the actual work. It removes the need to copy code back and forth or run separate sessions when pairing or reviewing.

How it works

The host presses a Share button (or types /share-session) to get a link. A teammate pastes that link into their own Claude Code session, and from then on both see the same conversation and tool output in real time. Guest prompts are executed by the host's machine; the guest's model is not called. By default, any guest request to edit files, run commands, or use the network asks the host for approval first, and a watch-only mode is available. The plugin works in Claude Desktop and the terminal, using a public room server by default or a server you host yourself. Sessions can also be moved to a cloud-hosted container so teammates can keep working while your computer is off.

Who it's for

Teams already using Claude Code who want to pair on real tasks, review code changes together, or share a context without duplicating setup. It suits developers working in the same codebase who need to see each other's live interactions with Claude, and who want to keep control over what the guest can do on the host machine.

Highlights

  • Share a live Claude session with a link; teammates join from their own Claude Code.
  • See prompts, streamed replies, tool cards, files, and supported previews in real time.
  • Host retains control: guest edit/run/network requests ask the host first, or use watch-only mode.
  • Optional teams add sign-in, live session lists, and invitations inside Claude Code.
  • Works in Claude Desktop and terminal; public server included, or host your own.
  • Move the session to a cloud container so work continues while the host's computer sleeps.

Related

  • Project
    Noelo-Lab/kuna

    Kuna is a Rust‑written binary decompiler designed for LLM agents. It’s fast (top of DecBench), offers CLI, web‑assembly, and Ghidra integrations, and provides an extensible, phase‑based pipeline with tunable options. The project aims for autonomous, agent‑driven refinement of decompilation quality.

  • Project
    DannyMac180/skills

    A GitHub repo offering plug‑in style “skills” that extend Claude Code or any skill‑aware AI agent with ready‑made ModSmith mods, multi‑agent workflow orchestration, task relaying, prompt‑ambition boosting, and artifact explanation with spaced‑repetition learning.

  • Project
    ROCm/HIPIFY

    HIPIFY provides command‑line tools (primarily `hipify-clang`) that automatically translate CUDA C++ code into portable HIP code, enabling the same source to run on AMD and NVIDIA GPUs. It also can generate Markdown or CSV documentation of the supported CUDA‑to‑HIP API mappings.

  • Project
    probelabs/probe

    Probe is a local, AST‑aware code‑search engine with a built‑in LLM agent. It lets AI assistants retrieve whole functions/classes from large repositories using Elasticsearch‑style queries, without any embedding index. Available via CLI, MCP server, or Node.js SDK, it supports 20+ languages and works fully offline.