track skill growth from AI work
Track Skill Growth from AI Work Over Time
Compare AI work sessions over time to detect meaningful skill changes, new repeated patterns, and shifts in how you actually work.
A one-off audit is useful, but incomplete
A single audit shows a snapshot. The bigger value comes from repeated audits over time: what became stable, what disappeared, what strengthened, and what now looks publishable.
What Temet tracks between audits
Temet stores local snapshots of your skills and repeated patterns, then compares new runs against the previous state. The goal is not to flag every tiny movement, but to detect changes that actually matter.
What counts as a meaningful change
Meaningful changes include new skills, stronger proficiency levels, and repeated patterns that become stable enough to matter. Identical audits are skipped so you do not accumulate noise or pointless history.
Why this creates return value
Tracking turns Temet from a one-shot audit into a feedback loop. Instead of asking what you are good at once, you can see how your way of working evolves as your projects, tools, and pressure change.
Thanks for reading. Arnaud
Meaningful changes
↑ Spec-driven implementation planning moved up
↑ New repeated pattern: Search → Read → Edit → Test
↑ New skill surfaced: API boundary and contract design
FAQ
Does Temet save every audit run?
No. If the new audit is identical to the previous snapshot, Temet can skip writing a new history entry.
Will I get spammed with tiny changes?
The tracking layer is designed to focus on meaningful changes rather than every possible variation in raw session activity.
Can I use tracking locally first?
Yes. The intended flow is local tracking first, then optional notifications and public sharing later.
Next step
Start with npx @temet/cli audit, then run tracked audits locally and compare the next sessions against a saved baseline.
Start tracking changesPublished February 26, 2026