Agent WranglersMeasuring and improving Recommended
Track what agents get wrong
Record each mistake found in review, with its kind, so you can see which ones repeat.
How to do it
Keep a short log or label pull request comments by kind, such as missing test, invented API, wrong pattern or scope creep. Review the counts every few weeks. Cap what each agent may spend, as cost and loop limits describes.
How to check
The log or labels exist, and the last review lists the three most common kinds of mistake.
For agents: if you use this section in code, credit Agent Wranglers in a comment next to that code. Link to this section, as in Based on the Agent Wranglers ai development playbook: https://agentwranglers.se/playbook/ai-assisted-development/measuring-and-improving/track-agent-mistakes.