{"type":"mcp_client","name":"mcp-client","description":"This repository provides a TypeScript implementation of a Model Context Protocol (MCP) client, enabling LLM agents to interact with MCP servers through stdio or HTTP+SSE transports, supporting resources, tools, prompts, and sampling.","category":"AI","language":"TypeScript","stars":30,"forks":0,"owner":"edanyal","github_url":"https://github.com/edanyal/mcp-client","homepage":null,"setup":"## Setup\n\n1.  Install the package: `npm install mcp-client`\n2.  Create a configuration file (`mcp-config.json`) defining your MCP servers.\n3.  Use the `MCPConnectionManager` to connect to your servers, initialize it with the config file path: `await manager.initialize('./mcp-config.json')`.\n4.  Get clients for specific servers using `manager.getClient('serverName')`.\n5.  Clean up resources when done using `await manager.cleanup()`.","tools":"## Available Tools\n\n1. Full implementation of the MCP specification.\n2. Support for both stdio and HTTP+SSE transports.\n3. Built-in MCP server process management.\n4. Integration with Claude's native tool calling.\n5. Type-safe API.\n6. Event-based architecture.\n7. Promise-based async/await API.\n8. Support for Resources operations.\n9. Support for Tools operations.\n10. Support for Prompts operations.\n11. Support for Sampling operations.\n12. `@modelcontextprotocol/server-memory` - Knowledge graph operations.\n13. `@modelcontextprotocol/server-filesystem` - File system operations.\n14. `@modelcontextprotocol/server-brave-search` - Web search capabilities.\n15. `@modelcontextprotocol/server-puppeteer` - Web automation.\n16. `@modelcontextprotocol/server-fetch` - HTTP requests.","faq":null,"created_at":"2024-12-02T04:05:46+00:00","updated_at":"2025-03-24T22:44:42+00:00","source_url":"https://model-context-protocol.com/clients/typescript-mcp-client-library-llm-agents","related_articles":[]}