The system that gets better every time you use it
One skill learns from the session you are currently working in. The other reads across your sessions and turns the findings into a skill or a rule.
What's inside
Retrospective Skill
After each piece of work, run /retrospective. Sub-agents scan your conversation for corrections, redone work, missing steps, and what worked well.
They propose updates to your skill files and new memories. You review, approve, and the next session is better.
/retrospective # analyze current session /retrospective video # focus on video skill
Full observability. Full control of what gets written to your memory.
Session Intelligence Skill
Session Intelligence reads a week or any window of Claude Code sessions from the local ~/.claude/projects folder.
It produces a ranked roadmap of recurring friction, skill improvements, and automation candidates. Every finding links back to evidence from the source session.
The findings become a skill or a rule, so the setup gets better across sessions instead of learning from only the current conversation.
How They Work Together
Retrospective learns from the session you are in. It proposes improvements while the corrections are still visible in the current context.
Session Intelligence reads across sessions. It surfaces what keeps repeating, where things keep breaking, and how you actually work.
Together, the findings become a skill or a rule and your setup gets better.
By signing up you'll receive the newsletter and occasional updates about new content and the Lab. Unsubscribe anytime.
Build the whole system
30-Day Agentic Second Brain
If you want the retrospective skill and the whole system built, the link is here.