This MCP server provides intelligent keyboard shortcut lookup using natural language queries powered by Claude Opus.
Key Capabilities:
Natural Language Querying: Search for keyboard shortcuts using conversational queries that Claude Opus intelligently matches against relevant shortcut data
Multi-Platform Support: Provides shortcuts for Ubuntu, macOS, and Windows across different operating systems
Desktop Environment Specificity: Includes shortcuts for desktop environments like GNOME and KDE
Application Coverage: Supports desktop apps (Firefox, VS Code), CLI tools (tmux, vim), and general system functions
Fast Performance: All shortcut data is loaded in-memory for quick filtering and retrieval
Claude Integration: Designed as an MCP server for Claude Desktop and Claude's Computer Use tool to enable automated computer interactions
Provides keyboard shortcuts for Firefox browser functionality including tab management, navigation, and browser-specific features
Provides keyboard shortcuts for GNOME desktop environment including window management, system functions, and desktop-specific operations
Provides keyboard shortcuts for KDE desktop environment operations and functionality
Provides keyboard shortcuts for macOS system operations, window management, and native applications
Provides keyboard shortcuts for tmux terminal multiplexer including pane management, window operations, and session control
Provides keyboard shortcuts for Ubuntu Linux system operations, applications, and desktop environment interactions
Provides keyboard shortcuts for Vim text editor including navigation, editing commands, and modal operations
Click on "Install 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., "@Keyboard Shortcuts MCP Server how do I switch between tabs in Firefox on Ubuntu?"
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.
A Model Context Protocol (MCP) server that provides keyboard shortcuts for various operating systems, desktop environments, and applications. Uses Claude Opus for intelligent natural language querying of shortcuts.
Primary Use Case: Designed to support Claude's Computer Use tool by providing accurate, context-aware keyboard shortcuts for automated computer interactions.
š Intelligent Search: Uses Claude Opus to understand natural language queries
š„ļø Multi-Platform: Support for Ubuntu (GNOME desktop), with architecture for macOS/Windows
š± App-Specific: Desktop apps (Firefox, VS Code), CLI tools (tmux, vim), and system shortcuts
š Fast: All shortcut data loaded in-memory for quick filtering
Related MCP server: Rebrain MCP Server
# Install dependencies
pnpm install
# Build the server
pnpm buildSet your Anthropic API key as an environment variable:
export ANTHROPIC_API_KEY="your-api-key-here"pnpm startAdd to your Claude Desktop MCP configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"keyboard-shortcuts": {
"command": "node",
"args": ["/path/to/keyboard-shortcuts-mcp/dist/index.js"],
"env": {
"ANTHROPIC_API_KEY": "your-api-key-here"
}
}
}
}get_shortcutsQuery keyboard shortcuts using natural language.
os (required): Operating system - "ubuntu", "macos", or "windows"
query (required): Natural language question about shortcuts
desktop (optional): Desktop environment - "gnome", "kde", etc.
application (optional): Specific application - "firefox", "tmux", etc.
// Query tmux shortcuts (no desktop needed for CLI tools)
{
"os": "ubuntu",
"application": "tmux",
"query": "how do I split a pane vertically?"
}
// Query Firefox shortcuts on GNOME
{
"os": "ubuntu",
"desktop": "gnome",
"application": "firefox",
"query": "how to open a new private window"
}
// Query GNOME window management
{
"os": "ubuntu",
"desktop": "gnome",
"query": "tile window to left half of screen"
}Shortcuts are organized as:
data/
āāā ubuntu/
āāā desktops/
ā āāā gnome/
ā āāā window-management.json
ā āāā system-functions.json
ā āāā ...
āāā apps/
ā āāā browser/
ā ā āāā firefox.json
ā ā āāā chrome.json
ā āāā editor/
ā ā āāā vscode.json
ā ā āāā vim.json
ā āāā ...
āāā tools/
āāā tmux.json
āāā vim.json
āāā ...Each JSON file contains:
{
"os": "ubuntu",
"desktop": "gnome" | null,
"application": "firefox" | null,
"file": "firefox",
"categories": [
{
"name": "Tab Management",
"shortcuts": [
{
"keys": "Ctrl + T",
"description": "New tab"
}
]
}
]
}# Watch mode (auto-reload on changes)
pnpm dev
# Build TypeScript
pnpm build
# Run built server
pnpm startData Loader (src/data-loader.ts): Loads all JSON files at startup into memory
Opus Client (src/opus-client.ts): Wraps Anthropic SDK for intelligent queries
MCP Server (src/index.ts): Exposes get_shortcuts tool via MCP protocol
The server filters shortcuts by OS/desktop/app, then passes relevant data + user query to Claude Opus for intelligent matching.
MIT
Query keyboard shortcuts for a specific OS, desktop environment, and/or application. Uses Claude Opus to intelligently search and return relevant shortcuts based on natural language queries.
| Name | Required | Description | Default |
|---|---|---|---|
| os | Yes | Operating system (e.g., ubuntu, macos, windows) | |
| query | Yes | Natural language query about keyboard shortcuts (e.g., "how do I split a pane vertically?") | |
| desktop | No | Desktop environment (e.g., gnome, kde) - optional | |
| application | No | Specific application (e.g., firefox, tmux) - optional |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It mentions using 'Claude Opus to intelligently search,' which adds some context about the AI-driven behavior, but it fails to disclose critical traits like rate limits, authentication needs, response format, or potential errors. For a tool with no annotation coverage, this is a significant gap in behavioral transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise and front-loaded, consisting of two sentences that efficiently convey the tool's purpose and method. Every sentence adds value without redundancy, though it could be slightly more structured (e.g., separating functional and behavioral aspects).
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (4 parameters, no annotations, no output schema), the description is moderately complete. It covers the core functionality and method but lacks details on output format, error handling, or operational constraints. Without an output schema, the description should ideally hint at return values, which it doesn't, leaving gaps in contextual understanding.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, meaning the input schema fully documents all parameters. The description adds minimal value beyond the schema by mentioning 'natural language queries' and 'intelligently search,' which aligns with the 'query' parameter but doesn't provide additional syntax or format details. The baseline score of 3 is appropriate given the high schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Query keyboard shortcuts for a specific OS, desktop environment, and/or application.' It specifies the verb 'query' and the resource 'keyboard shortcuts,' and mentions the intelligent search capability using Claude Opus. However, with no sibling tools, the differentiation aspect is not applicable, preventing a perfect score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage by stating it handles 'natural language queries' and targets specific OS/application contexts, but it lacks explicit guidance on when to use this tool versus alternatives (e.g., other search methods). With no sibling tools, there's no comparison, but it doesn't provide exclusions or prerequisites, leaving room for improvement.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
With only one tool, there is no possibility of ambiguity or confusion between tools. The single tool 'get_shortcuts' has a clear, distinct purpose that cannot be mistaken for any other functionality within this server.
The single tool name 'get_shortcuts' follows a clear verb_noun pattern (get + shortcuts). With only one tool, consistency is inherently perfect as there are no other names to compare or conflict with.
A single tool is generally too few for a server's purpose, especially one focused on keyboard shortcuts which could benefit from operations like listing available OS/applications, updating shortcuts, or searching by category. The scope feels thin and incomplete with just one query tool.
The server is severely incomplete for a keyboard shortcuts domain. While 'get_shortcuts' provides query functionality, there are obvious gaps: no ability to list supported systems/applications, create custom shortcuts, update existing ones, or manage shortcut sets. This will limit agent workflows significantly.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
MCP connector that lets ChatGPT list, search, and run your Apple Shortcuts via a local Mac agent
Search 200+ UnoRouter models (most free), check pricing, and chat through one key
Search your knowledge bases from any AI assistant using hybrid RAG.
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/jenova-marie/keyboard-shortcuts-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
