SkillAgentSearch skills...

Control-PC-Terminal

Control Your Desktop AI Agents From Any Device πŸš€

Install / Use

claude mcp add kumar045 -- npx -y github:kumar045/Control-PC-Terminal

If the server publishes to npm under a different name, use that package instead β€” check the repo README.

About this skill
πŸ”Œ

MCP Server

Model Context Protocol server

Quality Score

67/100

Supported Platforms

Claude Code
Claude Desktop
OpenAI Codex

Our assessment of Control-PC-Terminal

Control-PC-Terminal scores 67/100 on our quality scale, 908th of 963 AI & Machine Learning skills we index.

Its MCP Server is 4.8 KB long, well organised into 14 sections with 3 code examples: a solid amount of guidance for an agent.

It has 10 GitHub stars, so there is little community track record yet; judge it on its content.

Substance
26/30
Structure
18/20
Description
8/15
Adoption
4/20
Freshness
11/15

Maintenance, license and trust

  • The repository was last updated about 9 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
  • Our last check on 2026-10-22 found the source still online.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 86/100, with 2 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit β€” read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands (1 minor note below). An AI review of the same text found nothing harmful.

  • noteInstalls by piping a downloaded script into a shellline 42
    curl -fsSL https://raw.githubusercontent.com/kumar045/Control-PC-Terminal/main/install.sh | bash

AI review by kimi-k2.7-code on 2026-09-24. Automated pattern scan on 2026-09-24. It catches known dangerous patterns, not every risk β€” read a skill before letting an agent act on it.

Control-PC-Terminal compared with similar skills

All 4 of these similar skills score higher than Control-PC-Terminal; compare them before choosing.

SkillScoreStarsUpdatedFormat
Control-PC-Terminal (this skill)by kumar04567109mo agoMCP Server
claude-memby thedotmack100113.9ktodayCLAUDE.md
Agent-Reachby Panniantong100109.6ktodayCLAUDE.md
Understand-Anythingby Egonex-AI10091.1ktodayCLAUDE.md
headroomby headroomlabs-ai10079.0ktodayCLAUDE.md

Frequently asked questions

How do I install Control-PC-Terminal?
Run claude mcp add kumar045 -- npx -y github:kumar045/Control-PC-Terminal. The install tabs above show the steps for each supported agent.
Which AI agents does Control-PC-Terminal work with?
It is written for Claude Code, Claude Desktop and OpenAI Codex, as a MCP Server file. Other agents that read the same format can often use it too.
Is Control-PC-Terminal safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands (1 minor note below). An AI review of the same text found nothing harmful. It is MIT-licensed and scores 86/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is Control-PC-Terminal still maintained?
The repository was last updated about 9 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.

Control-PC-Terminal πŸš€

Control your Desktop AI Agents from anywhereβ€”via Telegram or Web Terminal.

πŸŽ₯ Watch the Demo Video

Control-PC-Terminal is an open-source tool that lets you control Claude Code, Codex, or any custom CLI agent running on your Desktop or VPS directly from your mobile phone.

Unlike standard remote tools, this is built for AI Agent persistence. Your agents run inside tmux sessions, so they keep working even if your internet drops. You can interact with them via a Secure Web Terminal or a Telegram Bot that streams logs and sends commands in real-time.

  • πŸš€ Ready-to-Run: Comes with preconfigured AI agents (using A2A, ADK, MCP, & Skills) so you can test them instantly!

  • 🌐 Language Agnostic: Build your own agents using ANY framework or language. Use the preconfigured ones as a blueprint.

  • πŸ”“ Total Freedom: Unlike ClawdBot (which locks you into their plugins), if it runs in a terminal, it works here!


✨ Key Features

1. πŸ“± Telegram Remote Control (New!)

Chat with your terminal! Control-PC-Terminal bridges Telegram messages directly to your running tmux session.

  • Send Prompts: Message the bot to send text directly to Claude/Codex.
  • Live Streaming: Use /watch to see the agent's output stream to your chat in near real-time.
  • Interrupts: Stop runaway agents instantly with /interrupt (sends Ctrl+C).
  • Control: Use /up, /down, /enter, and /yes to navigate menus remotely.

2. πŸ’» Persistent Web Terminal

  • tmux Integration: Your work keeps running on your machine even if your phone browser disconnects or screen locks.
  • ttyd Access: Turns your terminal into a web UI accessible from any browser (http://localhost:7681).
  • cloudflared Tunnel (Optional): Generates a secure public URL (e.g., trycloudflare.com) so you can access your terminal from outside your home network without router configuration.

πŸš€ Install

curl -fsSL https://raw.githubusercontent.com/kumar045/Control-PC-Terminal/main/install.sh | bash

The installer downloads the script and adds ~/.control-terminal/bin to your PATH.

πŸ–₯️ OS Support

  • Linux: Supported directly.
  • Windows: Use WSL (Windows Subsystem for Linux).

🧠 Usage

Run the tool:

control-terminal

You will be prompted to select an agent:

  1. codex
  2. claude
  3. other
  4. preconfigured custom agent

Telegram Setup

During startup, you can optionally configure Telegram control:

  1. Enter your Bot Token.
  2. Enter your Chat ID (for allowlist security).

Once running, the tool will:

  • Start the agent in a tmux session.
  • Launch the Telegram bot (if configured) to listen for commands.
  • Expose the session over a web browser via ttyd.

Telegram Commands

| Command | Description | | --- | --- | | /help | Show command list | | /status | Check if the tmux session is running | | /watch [s] | Stream terminal output to chat (default 2s interval) | | /unwatch | Stop live updates | | /tail [n] | Get the last n lines of output | | /interrupt | Send Ctrl+C to the agent | | /prompt <txt> | Send a prompt to the agent (or just type text) | | /yes / /no | Quick confirmation helpers | | /up / /down | Navigate command history/menus | | /enter | Send Enter key |

Note: If you send a plain text message, it is treated as a prompt and auto-enables /watch mode.


🧩 Why Control-Terminal is special

Control-Terminal combines three proven tools into one smooth workflow:

  1. tmux (The Engine): Keeps processes alive. If you use only a web terminal, your long-running agent dies when the tab closes. With tmux, it runs forever.
  2. ttyd (The Web View): Provides remote browser input/output to that specific tmux session.
  3. cloudflared (The Access): Gives you a shareable URL quickly, without manual router/NAT setup.

βš™οΈ Advanced Configuration

Preconfigured Custom Agents

You can define reusable agents in a config file so you don't have to type commands every time.

  • File location: ~/.control-terminal/custom-agents.conf
  • Format: name|command

Example:

a2a-adk-mcp|adk run --enable-a2a --enable-mcp --agent terminal_orchestrator

Credentials & Security

  • Web Auth: You can set a username/password for ttyd access during startup.
  • Telegram Auth: The bot only responds to the chat_id you provide.

Custom Script Requirements

If you are writing your own agent script (e.g., my_agent.py) to use with this:

  1. Read from stdin (terminal input).
  2. Write to stdout (terminal output).
  3. Keep running (don't exit after one output).

Control-Terminal handles the bridging between Telegram/Web and your script automatically.

Related Skills

View on GitHub
GitHub Stars10
CategoryAI
Updated8mo ago
Forks3

Languages

Shell

Trust signals

86/100

From repository metadata: license, adoption, age and documentation. Not a code audit β€” see the Safety scan above for what the skill file itself contains.

2 low1 info