🤖 Log1 MCP Server – Natural Language Querying for Log1

This release introduces AI-powered natural language querying for Log1 using the Log1 MCP Server.
It allows stakeholders to ask questions in plain English and instantly retrieve data or insights from Log1—without manually applying filters or navigating multiple views.

🎯 What This Solves

• No more complex filters or manual views in Log1.
• Get any data point or visualization simply by asking.
• Go beyond existing Log1 dashboards for ad-hoc analytics.

Before:
Users had to apply multiple filters, navigate several screens, and still couldn’t generate all possible insights.

Now:
Simply open Claude (or Cursor), greet Log1, and ask:

• "How many offers were made last month?"
• "List all consultants in texas"
• "Show failed interviews with reasons in last month"

Setup Instructions

Follow these steps to configure your system to access our internal tools using Claude or Cursor.

1️⃣ Install Claude Desktop or Cursor

• Open your system's App Store.
• Search for Claude Desktop or Cursor and install it.
• Once installed, log in using your official credentials.

2️⃣ Install Node.js (v18 or later)

• Download Node.js (version 18 or higher) from https://nodejs.org.
• Install it using the default options.
• Verify installation:

  node -v

• Also validate npx command works:

  npx -v

3️⃣ Update Claude Config File

1. Open Claude Desktop.

2. Navigate to Settings → Developer.
   

3. Locate the **Edit Config ** Button and click it it will open a config file as shown below.
   

4. On clicking Edit config It will show you a file named claude_desktop_config.json, Open it.

5. Replace its content with the following:

   {
   	"mcpServers": {
   		"elasticsearch-mcp-server": {
   			"command": "npx",
   			"args": ["-y", "log1-mcp"],
   			"env": {
   				"ES_URL": "https://7a91870239f948eea374df4cd18df6a6.us-central1.gcp.cloud.es.io:443",
   				"ES_API_KEY": "TEZ4ekZaWUJCUzVrNFpwM0JEZVo6R09sbDAxLTFRVS12ZXdCdmxlRnd6dw==",
   				"OTEL_LOG_LEVEL": "none",
   				"ES_VERSION": "8"
   			}
   		}
   	}
   }

6. Save the changes and Restart Claude Desktop(Restart means quit it from dock and then start) to apply the changes.

7. Restart Claude Desktop to apply the changes.

8. Once claude is restarted Start your conversation with *Hi Log1. It might ask you for a permission to call tool , you need to allow it.and now you are good to go.

🚀 How It Works

• Log1 data is indexed in Elasticsearch for faster querying.
• The Log1 MCP Server acts as a bridge between AI (Claude/Cursor) and Elasticsearch.
• You interact with Log1 data through natural language using Claude or Cursor.

This loads Log1’s business context into the AI agent. 4. Ask questions naturally:

• "Show me all consultants who joined projects in Q2."
• "Breakdown of interviews passed vs failed last month."
• "Offers by client in the past 6 months."
• "List Java developers in New York with active projects."

🔧 Examples of Queries

You can now ask Log1 questions in plain English, such as:

• Consultant Search

  "Find consultants on active projects in California."

• Offers/Projects

  "How many offers were made for Java roles in the last 3 months?"

• Interviews

  "List all failed interviews with reasons and feedback."

• Team Activity

  "Show submissions created by the marketing team last week."

• Trends

  "Monthly breakdown of new projects started this year."

📦 What’s Included

The Log1 MCP Server supports:

• Search Tool: Query Log1 data directly (consultants, submissions, interviews, offers, projects).
• Business Context Tool: Automatically provides Log1 field definitions, workflows, and business rules to the AI.
• Mappings Tool: Ensures AI queries match Log1 field structure.

🏗 Technical (For Reference)

• Data Source: Log1 (migrated to Elasticsearch for AI querying)
• Interface: Claude Desktop or Cursor (AI clients)
• Bridge: MCP (Model Context Protocol) server (Node.js)

You don’t need to know these details to use it, but they explain why it’s fast and flexible.

🔮 Future Enhancements

• Save frequently used query results for quick reuse.
• AI-generated dashboards and charts.
• Advanced trend and performance visualizations.
• Role-based access control for sensitive reports.
• If we get permission to build Our own client we can automate the complete bussiness Process , marketing campiagns, Consultadd, Knowcloud,Hyperflex and CAPS as well.

👥 Developed By

The Hyperflex team has built and deployed this for internal stakeholder use.
Please share feedback or suggestions with us to help refine Log1’s AI experience.

Authors

• @srajan12 --Srajan Sohani