Skip to content

How do I measure my Claude Code, Codex or Cursor use beyond tokens?

Short answer

Token and cost dashboards tell you how much the model was used. To understand how you worked with it, you need session-level measures: hands-on versus supervising time, sessions in flight, interruptions, failed tool calls and re-prompts. Ascenda Flow reads these from the history Claude Code, Codex, Cursor and Gemini CLI already keep.

Why it matters

Two weeks with the same token bill can be very different weeks: one with long, steady sessions, another full of failed runs and restarts. Tokens can't tell them apart. The shape of the work can.

How Ascenda helps

Flow captures live activity from Claude Code, Codex CLI, Cursor and Gemini CLI, and imports past history from Claude Code, Codex, Cursor and VS Code. From that it shows:

  • hands-on versus supervising time
  • sessions in flight at once, and project and context switching
  • autonomous agent runs and how often they stopped for you
  • context size and compactions
  • failed tool calls, re-prompts and retry storms
  • session shape, focus blocks and after-hours share

Because import comes first, your first read covers weeks you've already worked. After that Flow runs live.

Next step

See it on your own weeks

Flow is free for macOS. It imports the history your AI tools already keep, so the first read covers weeks you've already worked.