cli-mcp
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., "@cli-mcp Add 'git log' and show the last 5 commits"
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.
cli MCP server
cli bridge
Pretty much a remix of the demo notes app.
Components
Resources
The server implements a simple system for working with arbitrary clis with:
Custom cli:// URI scheme for accessing individual notes
Each cli resource has a name, help menu subtree, and text/plain mimetype
Tools
The server implements one tool:
add (cmd): Recursively parses cli help menu subtrees and stores definition
Takes "cmd"
Updates server state and notifies clients of resource changes
help (cmd): Return subtrees definition
run (cmd, cmd_args | None): Run a known command
Related MCP server: PyCLI MCP
Quickstart
Install
Claude Desktop
Note: I had to use /opt/homebrew/bin/uv instead of just uv in the command field in this json:
On MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%/Claude/claude_desktop_config.json
{"mcpServers": {
"cli": {
"command": "uv",
"args": [
"run",
"--directory",
"\<path to repo folder\>",
"cli"
]}
}
}By design, you have to explicitly 'add' a command before it can be run.
Debugging
Since MCP servers run over stdio, debugging can be challenging. For the best debugging experience, we strongly recommend using the MCP Inspector.
You can launch the MCP Inspector via npm with this command:
npx @modelcontextprotocol/inspector uv --directory <path to repo> run cliUpon launching, the Inspector will display a URL that you can access in your browser to begin debugging.
Available Tools
3 toolsTool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool has a distinct purpose: add learns a CLI schema, help retrieves it, and run executes a command. There is no overlap.
All tool names are single imperative verbs (add, help, run), following a consistent pattern.
Three tools is appropriate for a focused CLI schema management server, covering learning, retrieval, and execution.
Basic lifecycle is covered (add, retrieve, run), but lacks schema update or deletion, and may not support multiple CLIs.
Related MCP Connectors
Deploy, monitor, and manage your OpenClaw AI assistants via natural language.
Interact with the Stitch API using natural language commands.
Use your Mac, Windows or Linux computer from ChatGPT, Claude or Codex: files, commands, documents.
Source-checked CLI guides and model-aware planning for Claude Code, Codex, and Grok Build.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceTurn any CLI tool into an MCP server by leveraging its --help output, enabling natural language interaction with tools like gh, az, git, etc.16 npm25MIT
- AlicenseNot gradedqualityCmaintenanceEnables natural language interaction with any Python CLI application (Click, Typer, Argparse) through the Model Context Protocol.22MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to call any CLI tool by scanning its help output and serving it as an MCP server.1GPL 3.0
- AlicenseNot gradedqualityCmaintenanceUniversal MCP server that wraps any CLI tool, enabling AI assistants to run commands via natural language.1MIT