Loading...

The plugged.in MCP Proxy unifies multiple MCP servers, providing Knowledge, Memory, and Tools via a single interface for enhanced AI workflows and management.
Boost this tool
Subscribe to listing upgrades or segmented pushes.
The plugged.in MCP Proxy unifies multiple MCP servers, providing Knowledge, Memory, and Tools via a single interface for enhanced AI workflows and management.
The proxy is generally safe for read operations and controlled writes. Risks increase when destructive tools from connected servers are exposed without proper access controls. Secure configuration and monitoring are essential.
Performance depends on the number of connected MCP servers and the complexity of the AI workflows. Caching is used to improve discovery speed.
Cost depends on the usage of underlying MCP servers and any associated API costs. Consider the cost of document storage and processing for RAG.
npx -y @pluggedin/pluggedin-mcp-proxy@latest --pluggedin-api-key YOUR_API_KEY{
"mcpServers": {
"pluggedin": {
"command": "npx",
"args": ["-y", "@pluggedin/pluggedin-mcp-proxy@latest"],
"env": {
"PLUGGEDIN_API_KEY": "YOUR_API_KEY"
}
}
}
}PLUGGEDIN_API_KEYPLUGGEDIN_API_BASE_URLpluggedin_discover_toolsDiscovers available tools from connected MCP servers and built-in tools.
Read-only operation, no side effects.
pluggedin_ask_knowledge_baseSearches across documents using RAG with AI filtering capabilities.
Read-only operation, retrieves information from existing documents.
pluggedin_send_notificationSends notifications with optional email delivery.
Sends notifications, potential for spam if abused.
pluggedin_create_documentCreates AI-generated documents in the library.
Creates new documents, potential for content issues if not monitored.
pluggedin_clipboard_setSets a clipboard entry by name or index for persistent memory.
Stores data, potential for sensitive information exposure if not secured.
queryRuns read-only SQL queries on connected databases.
Read-only access to databases, no data modification.
API Key
hybrid
The proxy is generally safe for read operations and controlled writes. Risks increase when destructive tools from connected servers are exposed without proper access controls. Secure configuration and monitoring are essential.
Autonomy depends on the configuration of connected MCP servers and the permissions granted to the proxy. Exercise caution when enabling destructive tools.
Production Tip
Monitor the proxy's performance and resource usage to ensure optimal operation.
It's a middleware that aggregates multiple MCP servers into a single interface, providing unified access to Knowledge, Memory, and Tools.
It works with any MCP client, including Claude Desktop, Cline, and Cursor.
It supports API key authentication and optional Bearer token authentication for HTTP endpoints.
RAG v2 is the AI Document Exchange feature that allows MCP servers to create and search documents with full AI attribution.
Clipboard tools provide persistent memory for AI workflows, allowing you to store and retrieve data across sessions.
Use the `pluggedin_discover_tools` tool to discover available tools from connected MCP servers and built-in tools.
The proxy provides real-time notifications and activity tracking.