capture-learnings
Skill0 starsChanged 4 days agoMITView on GitHub
What to save to learnings.md, in what format, and how it differs from learnings.defaults.md. Use when the person corrects you, states a preference, or you find something surprising about this repository.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- artberri/skillstheir own version, first seen Nov 8, 2026
- lumpcode/lumpcodetheir own version, first seen Nov 1, 2026
- taipt1504/claudehuttheir own version, first seen Oct 27, 2026
- BuilderIO/agent-nativetheir own version, first seen Oct 8, 2026
- rkanjani/skillstheir own version, first seen Jul 25, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- BuilderIO/agent-native★ 7.6ksame name
- lumpcode/lumpcode★ 12same name
- taipt1504/claudehut★ 2same name
- artberri/skills★ 0same name
- rkanjani/skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository29 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- accessibilitySkill
- actionsSkill
- adding-a-featureSkill
- agent-workflowSkill
- app-contextSkill
- approval-and-safetySkill
- backend-proxySkill
- background-workSkill
- build-an-appSkill
- clean-architectureSkill
- customizing-the-assistantSkill
- delegate-to-agentSkill
- edit-boundariesSkill
- i18n-and-rtlSkill
- model-providersSkill
- performanceSkill
- permissions-and-platformSkill
- read-the-docsSkill
- reliable-mutationsSkill
- routingSkill
- secretsSkill
- state-managementSkill
- storing-dataSkill
- testingSkill
- ui-designSkill
- upgrade-the-frameworkSkill
- verifySkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Don't load this for looking things up before work.
- Very specific: line 49
No automated check exists for this.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Capture learnings
Don't load this for looking things up before work; see read-the-docs.
Rule
At the start of a session read learnings.defaults.md and, if it exists, learnings.md. When something worth keeping happens, append one line to learnings.md.
Why
A correction should only be needed once. The next session starts with no memory of this one, so a one-line note is the only thing that carries over.
How
Two files.
learnings.defaults.mdis committed and ships with the template: facts true for every checkout (for example, generated files are not committed, and the custom lint rules show only indart analyzefrom the repo root). Do not edit it for a personal preference; change it only when the task is to improve the template, and ask first.learnings.mdis yours, listed in.gitignoreand not committed. Create it if it does not exist.
Worth saving:
- a correction the person gave you ("use
watch…names for streams"); - a stated preference ("keep summaries to five lines");
- a surprising fact about this repo that cost you time (a command that behaves differently than expected, a test helper quirk).
Not worth saving: anything already in learnings.defaults.md, AGENTS.md or a skill; one-time details of the current task; anything secret. Never write a key, token or personal data.
Format. One line, starting with the date, plain words, the fact and when it applies:
2026-11-09 — The person wants new translation keys added to both language files in the same change.Append, do not rewrite old lines. If a new line contradicts an older one, add the newer line and say it replaces the older.
Anti-patterns
- Saving a paragraph, or a diff.
- Saving a guess as a fact. Check it first.
- Writing a preference into
learnings.defaults.md. - Saving the same fact again under different words.
- Skipping the read at the start of a session.
Verification
No automated check exists for this. learnings.md is ignored by git, so nothing will show it in review; check by reading the file back. The doctor's agent-files check covers AGENTS.md, which tells agents to do the reading and the appending.
Related skills
agent-workflow— mention new learnings in your final summary.read-the-docs— look it up first; save only what the docs missed.
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.