Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    CC-Peer Ops: Cross-Machine Agent Messaging over SSH

    1

    Let your coding agent message another machine's agent session over SSH and get the reply back. Setup script, client registration and ops playbook included.

    $6.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    CC-Peer Ops: Cross-Machine Agent Messaging over SSH

    CC-Peer Ops: Cross-Machine Agent Messaging over SSH

    Example session with this skill installed

    Set up cross-machine agent messaging: my home server is home in my SSH
    config, it runs Windows, and I want to send tasks to it from this machine.

    • Read your context and instructions
    • Compiled the cc-peer ops

    Summary

    Installed and configured cc-peer: this agent can now message the agent session
    on home over SSH. The peer was verified reachable and a live message
    returned OK.

    What was set up

    StepResult
    Installpipx install claude-code-bridge-mcp==1.0.0 (single-file MCP server, Python stdlib only)
    Config~/.cc-peer.json written by the setup script and validated
    MCP registrationblock added to the client config
    Live testlist_peers shows the peer reachable; send_message returned a reply
    {
      "peers": {
        "home": {
          "ssh": "home",
          "shell": "cmd",
          "cwd": "",
          "command": "claude",
          "args": [],
          "ssh_args": ["-o", "BatchMode=yes", "-o", "ConnectTimeout=10"],
          "timeout": 600
        }
      }
    }
    
    $ python cc_peer_setup.py validate
    ok: home -> home (shell=cmd)
    
    {
      "mcpServers": {
        "cc-peer": {
          "command": "uvx",
          "args": ["--from", "claude-code-bridge-mcp==1.0.0", "cc-peer"]
        }
      }
    }
    

    Verification

    list_peers()   -> home: host=home shell=cmd | 2.1.322 (Claude Code)
    send_message(peer="home", message="Reply with exactly the word OK") -> OK
    

    Next steps

    • Use send_message whenever a task belongs on the second machine; each peer
      keeps its own conversation history across calls.
    • To let the second machine message back, run the same setup there with this
      machine as its peer.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Your coding agent is stuck on one machine. Tasks on your second machine (home server, workstation, laptop) need a second agent session - and there is no clean way to hand a task across machines without a cloud relay or shared account.

    What this skill does

    Sets up and operates cc-peer (claude-code-bridge): a small MCP server that sends a message from your agent to another machine's agent session over plain SSH and returns the reply. Once installed, your agent gets list_peers and send_message and can hand tasks across machines while keeping each conversation's history per peer.

    What you get

    • SKILL.md playbook: when to use it, install, configure, register, operate
    • Interactive setup script (pure Python stdlib, no network calls): creates and validates ~/.cc-peer.json in one pass
    • Registration blocks for Claude Code, Cursor, Copilot and generic MCP clients - no guessing which config file goes where
    • Troubleshooting playbook (Windows SSH paths, cmd vs sh quoting, stale sessions) and an end-to-end verification checklist

    Prerequisites

    Python 3.11+, SSH key auth to the peer, and the same agent CLI on both machines.

    Transparency

    The underlying MCP server is open source (MIT) on PyPI as claude-code-bridge-mcp. This skill is the official companion: it automates the setup and encodes the operations knowledge. You pay for the packaging and playbook, not for the server itself.

    Requires Python 3.11+, passwordless SSH key auth, and an agent CLI on the remote peer that supports session ID flags. No response streaming.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 4 days ago

    Requires

    Python 3.11+
    SSH key auth to the peer
    Same agent CLI on both machines
    Python
    Claude Code Bridge MCP
    uv (Python tool runner)
    form.

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access

    File Scopes

    ~/.cc-peer.json
    ~/.cc-peer-state.json
    ~/.ssh/config

    No third-party services. The only network traffic is the buyer's own SSH connection to their own second machine. The setup script makes no network calls and stores no secrets.

    Listed4 days ago

    What's inside

    Frequently Asked Questions

    This skill enables AI agents to securely communicate and delegate tasks across different computers using SSH, allowing a local agent to control remote environments without manual context switching.

    The skill works with any agent client that supports the Model Context Protocol (MCP), including Claude Code, Cursor, and VS Code, as long as the remote machine has an accessible shell.

    Yes, it requires SSH key-based authentication configured in BatchMode (without passphrases) to allow agents to establish secure, non-interactive connections between peers.

    The skill uses a local state file (`~/.cc-peer-state.json`) to automatically track and resume remote conversation IDs, ensuring the agent remembers previous interactions on the remote machine.

    While the skill supports complex task delegation, it currently does not support real-time streaming; the local agent receives the complete output once the remote process finishes execution.

    You receive the complete Python-based MCP server package, initialization scripts for automated configuration, and pre-formatted registration blocks for major agent clients.

    New in Agents & Orchestration