kit-capture-lesson
Skill6 starsChanged 27 days agoMITView on GitHub
Turn a correction into a short written rule so it does not repeat. Use after steering, a repeated bug class, or “remember this”.
In this repository20 other files it gives its agents.
- AGENTS.mdAGENTS.md
- rules/AGENTS.mdAGENTS.md
- kit-audit-metricsSkill
- kit-audit-skillsSkill
- kit-challenge-designSkill
- kit-check-blastSkill
- kit-create-prSkill
- kit-debate-decision-treeSkill
- kit-debug-prodSkill
- kit-decisionsSkill
- kit-evolveSkill
- kit-explainSkill
- kit-learn-repoSkill
- kit-minimizeSkill
- kit-optimize-contextSkill
- kit-planSkill
- kit-proveSkill
- kit-reviewSkill
- kit-strip-diffSkill
- kit-triage-commentsSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 4 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Write the smallest rule that would have blocked this mistake.
- Very specific: line 13
Not domain terms or hard-to-reverse architecture (kit-challenge-design → CONTEXT / ADR)
- Very specific: line 14
Not a feature decision log (kit-decisions)
- Very specific: line 15
Not choosing a thinner implementation (kit-minimize)
- Very specific: line 16
Not a kit catalog audit (kit-audit-skills)
- Distinctive: line 17
Not a dump into CLAUDE.md / AGENTS.md / a new mega-skill
- Distinctive: line 32
Prefer patching an existing skill or rule over adding a name.
- Very specific: line 35
After yes: smallest edit. If you add or rename a skill/rule, update README.md catalogs and run kit-audit-skills.
- Very specific: line 41
Do not duplicate kit-plan, kit-prove, kit-minimize, kit-strip-diff, or kit-delegate
- Distinctive: line 42
Do not auto-apply. A proposed lesson that they reject is done.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Capture Lesson
Write the smallest rule that would have blocked this mistake. Apply only after explicit approval.
What this is NOT
- Not domain terms or hard-to-reverse architecture (
kit-challenge-design→ CONTEXT / ADR) - Not a feature decision log (
kit-decisions) - Not choosing a thinner implementation (
kit-minimize) - Not a kit catalog audit (
kit-audit-skills) - Not a dump into CLAUDE.md / AGENTS.md / a new mega-skill
Flow
- Failure - one sentence: what the agent did or missed.
- Rule - 1-5 lines that would have blocked it. Concrete trigger + action. No essays.
- Surface - pick one:
| Put it here | When |
| Existing skill | The miss belongs to a paced workflow already in the catalog |
Project rules/ | It must fire without a slash (style, git, safety, loop gates) |
| New skill | Rare: a full opt-in flow that is not an always-on habit |
| User-global memory | Only if they ask for a personal habit across repos |
Prefer patching an existing skill or rule over adding a name.
- Show the exact patch. Stop. Wait for yes.
- After yes: smallest edit. If you add or rename a skill/rule, update
README.mdcatalogs and runkit-audit-skills. - Approved or not: if
.kit-evolve/lessons.mdexists and this correction has no line yet, append one in its header's format.
Constraints
- Generic: no employer names, private hosts, or work-only tools (
AGENTS.md) - Do not duplicate
kit-plan,kit-prove,kit-minimize,kit-strip-diff, orkit-delegate - Do not auto-apply. A proposed lesson that they reject is done.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

