Skip to main content
Glama
MCP_INSPECTOR_CONNECTION.md2.1 kB
# Connecting MCP Inspector to Your Running Server ## 🔍 Current Setup - **MCP Inspector**: Running on port 6274 ✅ - **Your MCP Server**: Running on port 8003 ✅ - **URL Pattern**: `http://localhost:6274/?MCP_PROXY_AUTH_TOKEN=...#tools` ## 🔗 How to Connect ### Step 1: MCP Inspector Server Connection In your MCP Inspector interface: 1. **Look for "Add Server" or "Connect Server"** button 2. **Enter your server URL**: `http://localhost:8003` 3. **Transport**: Select "HTTP" 4. **No additional auth needed** - it's local ### Step 2: Expected URL Structure Once connected, your URL will be: ``` http://localhost:6274/?MCP_PROXY_AUTH_TOKEN=<token>&server=http://localhost:8003#tools ``` ### Step 3: What You Should See - ✅ **Server**: "Alfresco Document Management Server" - ✅ **15 Tools Available**: - search_content - search_by_metadata - advanced_search - cmis_search - browse_repository - repository_info - upload_document - download_document - checkout_document - checkin_document - cancel_checkout - delete_node - get_node_properties - update_node_properties - create_folder ## 🚨 If Connection Fails ### Check Server Status ```bash python tests/mcp_specific/test_server_status.py ``` ### Check Both Ports ```bash netstat -an | findstr ":6274\|:8003" ``` ### Restart If Needed ```bash # Stop your MCP server taskkill /PID 57420 /F # Restart with explicit logging fastmcp run alfresco_mcp_server/fastmcp_server.py --transport http --host localhost --port 8003 --log-level DEBUG ``` ## 🎯 Quick Test 1. **Open your MCP Inspector**: `http://localhost:6274/?MCP_PROXY_AUTH_TOKEN=4b4a857ba6eecd2ac7029bf0c9daaf97e311b7f5cee35dec3a25ddf739ffd66b#tools` 2. **Add server**: `http://localhost:8003` 3. **Test a tool**: Try "browse_repository" or "search_content" ## 🔒 The "Risky Auth" Token The `MCP_PROXY_AUTH_TOKEN=4b4a857ba6eecd2ac7029bf0c9daaf97e311b7f5cee35dec3a25ddf739ffd66b` is: - ✅ **Normal**: Generated by MCP Inspector for security - ✅ **Local only**: Not transmitted anywhere - ✅ **Safe**: Standard MCP development practice

Latest Blog Posts

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/stevereiner/python-alfresco-mcp-server'

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