study-me
Skill2 starsChanged 6 months agoMITView on GitHub
Freely study knowledge already saved in the current repo by extracting as many content-based questions as each note section supports and asking them one at a time, with no grading or scheduling — just self-check against the original text. Use when the user wants to freely study, self-test, or quiz themselves on notes without spaced-repetition tracking, or says "study me", "자유 학습", "편하게 공부해볼게".
In this repository17 other files it gives its agents.
- create-nextjs-specificationSkill
- create-product-demoSkill
- find-github-repoSkill
- find-npm-packageSkill
- image-styleSkill
- nextjs-route-scaffoldSkill
- perfect-promptSkill
- place-meSkill
- review-meSkill
- setup-aeo-geoSkill
- setup-harnessSkill
- songwritingSkill
- step-by-stepSkill
- stock-analysisSkill
- teach-meSkill
- three-linesSkill
- visualstormingSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 7 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
If I gave a topic/file as context → study exactly that.
- Distinctive: line 13
If I gave no context:
- Very specific: line 14
Read levels.md and propose the weakest topics first (Blank/Glimpsed before Grounded before Fluent).
- Very specific: line 15
If levels.md doesn't exist → ask me once which topic, folder, or file pattern to study.
- Distinctive: line 17
Don't queue the entire knowledge base by default — a growing KB makes that unusable.
- Distinctive: line 34
When I answer, reveal the relevant section content as the self-check reference — no verdict, no grade, no "correct/incorrect" label.
- Distinctive: line 35
Wait for me to say "다음"/"next" (or similar) before asking the next question.
- Distinctive: line 40
When the queue is empty, or I say to stop, give a short summary: how many questions covered, out of how many sections.
- Very specific: line 41
Never modify levels.md — study-me has no grading signal to base a level update on.
- Distinctive: line 42
Never modify the knowledge notes themselves — study-me only reads them.
- Very specific: line 43
No session state is persisted anywhere.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Freely study knowledge already saved in this repository. Companion to teach-me and review-me: teach-me writes notes, review-me quizzes with spaced-repetition grading, study-me quizzes with no grading at all — pure self-check, stop and resume freely.
Scope
Decide what to study before building the queue — same logic as review-me:
- If I gave a topic/file as context → study exactly that.
- If I gave no context:
- Read
levels.mdand propose the weakest topics first (Blank/Glimpsed before Grounded before Fluent). Ask: "Study [weak topics] today?" Stop and wait. - If
levels.mddoesn't exist → ask me once which topic, folder, or file pattern to study. Stop and wait.
- Read
Don't queue the entire knowledge base by default — a growing KB makes that unusable.
Targets
Knowledge notes live under the notes/ folder (any depth): every .md in notes/ within scope is a note. Each ## section inside a matched note is a source for questions.
Building the question queue
For each ## section in scope:
- Extract as many content-based questions as the section's content supports — no upper limit. A short section (one fact) yields one question; a dense section yields several, one per distinct fact or concept inside it.
- Base every question on the content itself, never just the section heading.
- Order: sections in file order; files in the order they appear in
_data/notes_order.ymlwhen the scope falls undernotes/(fallback: filesystem order if that file doesn't cover the scope). Within a section, ask questions in the order the underlying facts appear in the text.
Study loop
- Pick the next question from the queue. Ask it. Stop and wait.
- When I answer, reveal the relevant section content as the self-check reference — no verdict, no grade, no "correct/incorrect" label. Just the source text so I can compare it against my own answer.
- Wait for me to say "다음"/"next" (or similar) before asking the next question. Never ask two questions at once.
- Repeat until the queue is empty or I say I want to stop.
Ending
- When the queue is empty, or I say to stop, give a short summary: how many questions covered, out of how many sections.
- Never modify
levels.md— study-me has no grading signal to base a level update on. Level tracking belongs to place-me/review-me. - Never modify the knowledge notes themselves — study-me only reads them.
- No session state is persisted anywhere. If the session is interrupted, the next
study-merun simply starts a fresh queue — there is no dot-folder, no resume.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.

