# aidevelopers2/remoteopenclaw-mcp [Health: Active]

**Category:** 🔗 Aggregators  
**Repository:** https://github.com/aidevelopers2/remoteopenclaw-mcp  
**GitHub Stars:** 3  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/aidevelopers2-remoteopenclaw-mcp

## Description
MCP server and CLI to search the Remote OpenClaw directory of 13,870+ MCP servers, 4,384+ agent skills, and plugins. Returns names, links, and install commands. Install: claude mcp add remoteopenclaw -- npx -y remoteopenclaw. CLI: npx remoteopenclaw search . Free, no API key.

## Tools
Capabilities this server exposes over MCP:

- **search_mcp_servers** — Find an MCP server for a task (databases, scraping, GitHub…)
- **search_skills** — Find a reusable agent skill, with its install command
- **search_plugins** — Find a plugin / marketplace package

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "remoteopenclaw-mcp": {
    "command": "npx",
    "args": ["-y","remoteopenclaw"]
  }
}
```

## Documentation

## What aidevelopers2/remoteopenclaw-mcp MCP server does

aidevelopers2/remoteopenclaw-mcp MCP server exposes search access to the Remote OpenClaw directory through three MCP tools. The directory contains MCP servers, agent skills, and plugins. An agent can search for a capability such as database access, scraping, GitHub tooling, PDF-related skills, or testing plugins and receive matching directory entries.

The server is intended for tool discovery rather than execution. Its responses provide result metadata that can help an agent choose a project, including names, descriptions, links, popularity information, and installation commands. The documented directory size is more than 13,870 MCP servers and more than 4,384 agent skills, in addition to plugins.

## How it works

The package acts as a thin MCP wrapper around the public Remote OpenClaw search endpoint at `https://www.remoteopenclaw.com/api/search`. When an MCP client invokes a search tool, the server sends the request to that API and returns the matching directory results. Because the lookup occurs against the remote API, the package does not need to bundle a local copy of the directory.

The three exposed tools are:

- `search_mcp_servers`: searches for MCP servers related to a task.
- `search_skills`: searches for reusable agent skills and their installation commands.
- `search_plugins`: searches for plugins or marketplace packages.

Results are ranked by stars or installs according to the README. The project is built with the TypeScript Model Context Protocol SDK.

## Setup and configuration

Install the server as an MCP stdio process using the `remoteopenclaw` npm package. A client configuration has the following executable and arguments:

```json
{
  "mcpServers": {
    "remoteopenclaw": {
      "command": "npx",
      "args": ["-y", "remoteopenclaw"]
    }
  }
}
```

The documented Claude Code setup can also be added with `claude mcp add remoteopenclaw -- npx -y remoteopenclaw`. Cursor, Windsurf, and other MCP clients can use the same command-and-arguments configuration. No API key is required.

The package also provides a command-line search entry point: `npx remoteopenclaw search`. The provided material does not document additional flags or configuration variables.

## Tools and capabilities

For developers evaluating aidevelopers2/remoteopenclaw-mcp, the main capability is directory search through an agent interface. Natural-language requests can be mapped to the appropriate tool, allowing an agent to locate candidate integrations and present their links and setup commands. Searches cover three separate content types, which keeps MCP server, skill, and plugin discovery distinct.

This server can help an agent answer questions such as which MCP server supports a particular database, whether a reusable skill exists for a task, or what plugins are listed for testing. It does not provide the discovered servers' functionality itself; it returns information about those entries.

## Limitations and notes

The server depends on the public Remote OpenClaw search API, so search results require access to that endpoint. The supplied documentation does not describe offline behavior, authentication options, rate limits, search syntax, or response pagination. Popularity ranking is based on stars or installs, but no further ranking controls are documented.

The project is licensed under MIT. Its documented role is discovery: installing a result found through aidevelopers2/remoteopenclaw-mcp is a separate step from running the search server.

_Full upstream README: https://allmcps.com/mcp/aidevelopers2-remoteopenclaw-mcp/readme_