Skip to main content
Glama

Context Pods

by conorluddy
MCP_CLIENT_SETUP.mdโ€ข4.8 kB
# MCP Client Configuration Guide This guide explains how to connect to the Context-Pods Meta-MCP Server from various MCP clients. ## Overview The Context-Pods Meta-MCP Server exposes the Context-Pods toolkit as an MCP protocol interface, allowing AI systems to create and manage MCP servers programmatically. ## Claude Desktop Configuration Add this configuration to your Claude Desktop settings file: ### macOS Location: `~/Library/Application Support/Claude/claude_desktop_config.json` ### Windows Location: `%APPDATA%/Claude/claude_desktop_config.json` ### Configuration ```json { "mcpServers": { "context-pods": { "command": "npx", "args": ["@context-pods/server"], "env": { "NODE_ENV": "production" } } } } ``` ### Development Configuration For local development, use this configuration: ```json { "mcpServers": { "context-pods-dev": { "command": "node", "args": ["./packages/server/dist/index.js"], "cwd": "/path/to/MCPTemplate", "env": { "NODE_ENV": "development", "DEBUG": "context-pods:*" } } } } ``` ## Available Tools Once connected, the following MCP tools will be available: ### Core Generation Tools - **`create-mcp`** - Generate new MCP servers from templates - **`wrap-script`** - Convert existing scripts to MCP servers - **`list-mcps`** - Show all managed MCP servers - **`validate-mcp`** - Validate MCP servers against standards ### Example Usage ``` Human: I need an MCP server that can interact with a PostgreSQL database Claude: I'll help you create a PostgreSQL MCP server using Context-Pods! *Claude calls the create-mcp tool* Tool: create-mcp Arguments: { "name": "postgresql-server", "description": "MCP server for PostgreSQL database operations", "language": "typescript" } Result: โœ… Successfully created postgresql-server at ./postgresql-server ``` ## Available Resources The Meta-MCP Server provides these resources: - **`context-pods://templates/`** - List of available templates - **`context-pods://mcps/`** - All managed MCP servers - **`context-pods://status`** - System status and health - **`context-pods://statistics`** - Usage statistics ## Alternative MCP Clients ### Cody (VS Code) Add to VS Code settings.json: ```json { "cody.experimental.mcp": { "servers": { "context-pods": { "command": "npx", "args": ["@context-pods/server"] } } } } ``` ### Continue (VS Code) Add to continue configuration: ```json { "mcpServers": [ { "name": "context-pods", "command": "npx", "args": ["@context-pods/server"] } ] } ``` ### Custom Integration For custom integrations, use the MCP SDK to connect: ```typescript import { Client } from '@modelcontextprotocol/sdk/client/index.js'; import { StdioClientTransport } from '@modelcontextprotocol/sdk/client/stdio.js'; const transport = new StdioClientTransport({ command: 'npx', args: ['@context-pods/server'] }); const client = new Client({ name: 'my-client', version: '1.0.0' }, { capabilities: {} }); await client.connect(transport); // List available tools const tools = await client.listTools(); console.log('Available tools:', tools.tools); // Call the create-mcp tool const result = await client.callTool({ name: 'create-mcp', arguments: { name: 'my-server', description: 'My custom MCP server' } }); ``` ## Troubleshooting ### Common Issues 1. **Server not starting** - Ensure Node.js 18+ is installed - Run `npm install` in the Context-Pods directory - Check that `@context-pods/server` is built: `npm run build` 2. **Permission errors** - Ensure the client has permission to execute Node.js - Check file permissions on the server directory 3. **Tools not appearing** - Restart your MCP client - Check the client logs for connection errors - Verify the configuration file syntax ### Debug Mode Enable debug logging by setting environment variables: ```json { "mcpServers": { "context-pods": { "command": "npx", "args": ["@context-pods/server"], "env": { "DEBUG": "context-pods:*", "NODE_ENV": "development" } } } } ``` ### Log Locations - **Claude Desktop**: Check the application logs - **Development**: Logs appear in the terminal where the server is running - **Production**: Logs are written to the system log ## Next Steps 1. Install and configure your MCP client 2. Test the connection by listing available tools 3. Try creating your first MCP server with the `create-mcp` tool 4. Explore the available templates and resources For more information, see: - [Templates Guide](./TEMPLATES.md) - [Server Architecture](./ARCHITECTURE.md) - [Contributing Guide](../CONTRIBUTING.md)

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/conorluddy/ContextPods'

If you have feedback or need assistance with the MCP directory API, please join our Discord server