Inspiration

As a developer, I spend a lot of time debugging production logs. It's slow and stressful. I wanted to build something that makes this faster.

What it does

You paste a log or stacktrace, LogWise analyzes it using Splunk MCP and AI, and gives you root cause, similar past incidents, incident timeline, resolution time prediction, action plan, and a PDF report.

How we built it

React + Spring Boot backend + Groq AI + Splunk MCP Server. Splunk is used to store and query incident history in real time via JSON-RPC.

Challenges we ran into

The biggest challenge was connecting to Splunk MCP — port 8089 is blocked on Splunk Cloud Trial. I found an alternative endpoint on port 443 that made it work. Also had SSL issues with Java that I fixed with a custom WebClient configuration.

Accomplishments that we're proud of

Successfully connected to the real Splunk MCP Server via JSON-RPC and used it for 3 distinct operations. Built 11 working features in under 2 weeks.

What we learned

How to build an agentic workflow combining Splunk observability with AI. And how to overcome Splunk Cloud Trial limitations to make a real MCP integration work.

What's next for LogWise

Auto-connect to live log streams, Slack/Teams alerts for critical incidents, and a team dashboard to track recurring errors across the organization.

Built With

Share this project:

Updates