Skip to main content
Version: 6.1

AI Observability: Claude Code

The AI Observability: Claude Code dashboard is designed to monitor local Claude Code activity.
It displays dialog activity, token consumption, application events, and usage distribution by users and projects.

Useful for: operations engineers, AI team leads, usage analysts.

Data source: gen_ai_events*

Main Sections​

1. Claude Code Activity​

Shows summary activity of Claude Code for the selected period:

  • number of working projects, completed requests, events, and application errors
  • dynamics of events, requests, models, and entry points over time

Claude Code Description and Activity

2. Claude Code Dialogs​

Shows a table of Claude Code dialogs and details of the selected interaction:

  • request execution time
  • user and entry point
  • request and response text
  • provider, model, and token consumption
  • drilldown panel with details of the selected dialog

When selecting a dialog, the Claude Code Dialog Details panel opens, displaying request parameters, tokens, model, user, entry point, question text, and the full Claude response.

Claude Code Dialogs

3. Tokens, Users, and Projects​

Shows Claude Code token consumption broken down by time, users, and working projects:

  • total token consumption
  • input and output tokens
  • token distribution over time
  • token consumption by users
  • token consumption by projects

Tokens, Users, and Projects

4. Application Events and Errors​

Shows technical events of local Claude Code and errors recorded in application logs:

  • application events by messages
  • application events by log files
  • number of events by logger sources
  • application errors and related messages

Application Events and Errors