PC Bridge Linux
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@PC Bridge LinuxShow me the git status and recent commits in my workspace"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
PC Bridge Linux
Connect an MCP-compatible AI client to your Linux computer to work with local files, commands, and projects. Includes a per-user installer and guides for connecting ChatGPT, Unity, and Blender.
Unity and Blender integrations are included; the editors themselves are not installed
automatically. Blender support includes the addon and an optional MCP server environment
installed with --with-blender; you must install Blender and enable the addon manually.
Unity support includes MCP setup instructions and a package configuration reference;
you must install Unity and enable its MCP plugin in your project.
See the Unity and Blender setup guide.
By SurveysGuy · X: @someguy_112358
MCP (Model Context Protocol) lets an AI client discover and call the Bridge's tools. This project does not include an AI model. You need an MCP client and, for remote access from ChatGPT, your own account and tunnel connection.
What it does
Provides 20 MCP tools for files, search, shell commands, Git, system diagnostics, screenshots, and authorized application management.
Runs shell, Git, and patch operations through Bubblewrap, constrained by the configured workspace and filesystem policy, with networking disabled inside the command sandbox.
Supports desktop screenshots through Wayland portals or X11, subject to desktop permissions.
Optionally installs a separate Python environment for Blender MCP, with its addon included.
Documents Unity MCP setup inside your project and how to connect all three servers. The Bridge handles files and commands; the editors own their live state.
Related MCP server: sshand
Why use it?
Use it when you want an assistant to work on Linux projects through explicit tools, without repeatedly copying files and terminal output into a conversation. Examples include investigating a project error, editing files in an authorized folder, checking Git status, or combining code work with Unity and Blender tools.
The installer uses an independent directory and dedicated Python environments. It preserves existing configuration when run again, making it easier to try the integration alongside an existing setup and remove it later.
Your workspace is the folder you authorize the assistant to use. Its tools can modify and delete files there: choose the projects you want to expose and configure protected paths. The sandbox is a development boundary, not containment against hostile processes running under the same local account. Editor MCPs have their own permissions, including code execution inside the editor.
Compatibility
Tested on Ubuntu 26.04 LTS with Python 3.14.4 and systemd 259. Other Linux distributions may work if they meet the requirements, but have not been validated. Requires Python 3.12+, venv, Git, and working Bubblewrap. Application launching requires a compatible user systemd manager; screenshots require a graphical session. The installer rejects Windows, macOS, and WSL.
Quick start
On Ubuntu/Debian with Python 3.12 or newer, install the system dependencies and clone into a permanent directory owned by your user:
sudo apt install python3 python3-venv git bubblewrap
git clone https://github.com/SomeGuySomewhereSometime/pc-bridge-linux.git PCBridgeLinux
cd PCBridgeLinux
sh install.sh --check
sh install.sh --workspace "$HOME/PCBridgeLinuxProjects"
.venv/bin/python run_bridge.py doctorUse sudo only for the system package manager, not for the installer.
Add --with-blender to install the separate Blender MCP Python environment.
Installation downloads the pinned Python dependencies inherited from the base project.
Full installation guide · Unity and Blender · MCP clients and ChatGPT · Validation and limitations
The installer generates .local/mcp-client.json with absolute paths for clients
that accept mcpServers. You do not need to activate the virtual environment.
Account/tunnel setup and editor addon activation are manual steps covered in the guides.
Scope
Native Linux, with Wayland or X11 for desktop features.
A working Bubblewrap sandbox is mandatory for a full installation.
Wayland screenshots depend on the desktop portal and user consent; X11 uses MSS.
Editors can be opened manually when compatible user systemd is unavailable.
Existing configuration is preserved. Projects are kept outside the installation directory.
The installer does not create services or change other installations, accounts, or projects.
Provenance and licensing
Derived from the local PCBridgePortable source snapshot dated 2026-09-08. Changes focus on the Linux installer, entry point, documentation, and CI. Shared internal modules retain adapters and tests from the base; they do not imply Windows support in this distribution.
The Blender addon retains its original MIT license. The base project does not specify a redistribution license for the Bridge code; this repository does not add one.
Build a distribution archive
Run sh package.sh to create PCBridgeLinux.tar.gz and its SHA-256 file in the
parent directory. The archive includes source, documentation, and the addon,
while excluding virtual environments, personal configuration, .local, and Git history.
Share that archive rather than a configured installation directory.
This server cannot be deployed
Maintenance
Related MCP Connectors
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
Unified MCP Server is a remote MCP connector for AI agents and vertical AI products that provides access to 22,000+ authorized SaaS tools across 400+ integrations and 24 categories directly inside LLMs (Claude, GPT, Gemini, Cohere). Tools operate only on explicitly authorized customer connections, enabling agents to safely read and write against live third-party systems.
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Hosted MCP server connecting claude.ai, ChatGPT and other AI apps to your own computer
Related MCP Servers
- AlicenseNot gradedqualityAmaintenanceExposes full Linux system control to AI clients via MCP tools for bash execution, file operations, and search, with configurable permissions and audit logging.2Apache 2.0
- AlicenseAqualityAmaintenanceAn open MCP server that gives any AI agent SSH access to remote Linux/Unix machines — shell commands, file read/write, and SFTP transfers.11MIT
- AlicenseNot gradedqualityCmaintenanceA secure MCP server that exposes local filesystem operations to AI clients with sandboxed access and runtime directory changes.1MIT
- AlicenseAqualityBmaintenanceSelf-hosted MCP server that gives AI assistants a real Linux development environment: file editing, shell commands, tmux terminals, and browser screenshots, with bounded access through a project registry and write gates.39MIT