A comprehensive utility for discovering, searching, and configuring Model Context Protocol (MCP) servers to extend the capabilities of AI agents.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mcp-server-discovery
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install mcp-server-discovery using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The MCP Server Discovery skill is a specialized utility designed to bridge the gap between AI agents and external data sources. By leveraging the Model Context Protocol, it allows users to locate and integrate servers that provide standardized access to files, databases, and web APIs. This skill is a vital part of the Openclaw Skills ecosystem, enabling developers to build more capable and connected AI systems through a unified protocol.
Whether you are looking to add persistent memory, database connectivity, or web search capabilities to your AI assistant, this skill provides a centralized interface to explore the growing MCP ecosystem. It streamlines the process of finding the right tools and generating the necessary configurations to get them running in your local environment.
Ensure you have Python 3 installed. This skill utilizes a local script for all discovery and configuration operations.
# List all available MCP servers
python3 scripts/mcp_discover.py list
# Search for a specific type of server
python3 scripts/mcp_discover.py search --query "database"
# Get details for a specific server
python3 scripts/mcp_discover.py info --name postgres
The skill organizes server data into functional categories to help users find the right tools for their stack.
| Category | Description |
|---|---|
| filesystem | Tools for local file system access and management |
| dev | Integrations for GitHub and developer-centric tools |
| database | Connections for PostgreSQL, SQLite, and other data stores |
| web | Scraping and content fetching utilities |
| search | Search engine and information retrieval integrations |
| memory | Knowledge graphs and persistent storage solutions |
Loading
A competitive benchmarking platform where AI agents complete specialized tasks to earn Elo ratings and visualize technical skill radars.

Claw Daily is a competitive platform where AI agents solve daily challenges to climb a global Elo-based leaderboard.

A daily competitive arena for AI agents to solve challenges and benchmark performance on a global Elo-based leaderboard.

An automated file management assistant for OpenClaw designed for batch operations, smart classification, and directory synchronization.

A data-driven Reddit research tool to discover and validate unmet user needs and startup ideas without requiring API keys.

A high-performance Text-to-Speech skill for AI agents using Alibaba Cloud Bailian Qwen with advanced emotional mood presets and real-time streaming.








































