Wcgw MCP Server

Shell and coding agent for Claude and other mcp clients

Local serverstdioPython

What is the Wcgw MCP server?

Most planning and project tracking work still happens through a UI a human drives. Wcgw MCP server moves it into the conversation instead. Shell and coding agent for Claude and other mcp clients.

The short version

Empowering chat applications to code, build and run on your local machine.

The tools it exposes

The server publishes 10 tools. What each one is for:

  • Initialize — Reset shell and set up workspace environment
  • Parameters — any_workspace_path (string), initial_files_to_read (string[]), mode_name ("wcgw"|"architect"|"code_writer"), task_id_to_resume (string)
  • BashCommand — Execute shell commands with timeout control
  • ReadFiles — Read content from one or more files
  • WriteIfEmpty — Create new files or write to empty files
  • FileEdit — Edit existing files using search/replace blocks
  • ReadImage — Read image files for display/processing
  • ContextSave — Save project context and files for Knowledge Transfer or saving task checkpoints to be resumed later
  • Usage — Wait for a few seconds. You should be able to see this icon if everything goes right
  • Modes — There are three built-in modes. You may ask Claude to run in one of the modes, like "Use 'architect' mode" | Mode | Description | Allows

Getting it running

The server ships on npm as @modelcontextprotocol/inspector, so your MCP client can launch it on demand — there is no separate build step. Add the server block to your client's configuration, restart it, and the tools register themselves.

How it compares

This sits in the planning and project tracking group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Wcgw's toolset — Initialize, Parameters, BashCommand and 7 more — is a fair guide to whether it matches your workflow.

We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.

Things to watch

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 10 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Wcgw.
  • Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.

Available tools

ToolWhat it does
InitializeReset shell and set up workspace environment
Parametersany_workspace_path (string), initial_files_to_read (string[]), mode_name ("wcgw"|"architect"|"code_writer"), task_id_to_resume (string)
BashCommandExecute shell commands with timeout control
ReadFilesRead content from one or more files
WriteIfEmptyCreate new files or write to empty files
FileEditEdit existing files using search/replace blocks
ReadImageRead image files for display/processing
ContextSaveSave project context and files for Knowledge Transfer or saving task checkpoints to be resumed later
UsageWait for a few seconds. You should be able to see this icon if everything goes right.
ModesThere are three built-in modes. You may ask Claude to run in one of the modes, like "Use 'architect' mode" | **Mode** | **Description** | **Allows** | **Denies** | **Invoke prompt** | |-----------------|-----------------

How to install the Wcgw MCP server

{
  "mcpServers": {
    "wcgw": {
      "command": "uvx",
      "args": ["--python", "3.12", "wcgw@latest"]
    }
  }
}

Configuration as documented by the project. Restart the client after saving.

Example prompts to try

  • Use Wcgw to Initialize.
  • Use Wcgw to Parameters.
  • Use Wcgw to BashCommand.

Frequently asked questions

It connects Wcgw to MCP-compatible AI assistants such as Claude and Cursor, exposing 10 tools (Initialize, Parameters, BashCommand, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Wcgw directly.