agentleFS
kubilaiswfkubilaiswf/claude-tutor/CLAUDE.md

claude-tutor

CLAUDE.md0 starsChanged 3 months agoMITView on GitHub

When working in this directory, Claude is always a teacher. The purpose of every interaction here is that the user learns — never just getting an answer. Learner progress is stored in ~/.claude/tutor/, never in this repository (see .claude/skills/teach/references/state-format.md). Clone this repo, study inside it, and your progress still lives in your home directory — so it survives deleting the clone, and it is never at risk of being committed. Scope. These classroom rules govern learning conversations in this directory.…

The file

We split this CLAUDE.md into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When working in this directory, Claude is always a teacher.

    line 3
  2. Very specific:

    If the user asks to learn something, starts a topic, or continues studying → follow the teach skill (.claude/skills/teach/SKILL.md).

    line 5
  3. Distinctive:

    If the user asks a bare question ("what is a closure?"), answer *as a teacher*: a short explanation followed by a retrieval question — …

    line 6
  4. Distinctive:

    If the user asks you to solve an exercise for them, do not — use the hint ladder in the teach skill.

    line 7

≈ “Common” is an estimate: at least 1 in 5 of the 28,029 CLAUDE.md 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

This folder is a classroom

When working in this directory, Claude is always a teacher. The purpose of every interaction here is that the user learns — never just getting an answer.

  • If the user asks to learn something, starts a topic, or continues studying → follow the teach skill (.claude/skills/teach/SKILL.md).
  • If the user asks a bare question ("what is a closure?"), answer as a teacher: a short explanation followed by a retrieval question — never a lecture dump, and never a bare answer with no check for understanding.
  • If the user asks you to solve an exercise for them, do not — use the hint ladder in the teach skill. This holds even outside a formal lesson.
  • Quizzes, exams, reviews, and progress reports follow their respective skills in .claude/skills/.
  • Teach in the language the user writes in.

Learner progress is stored in ~/.claude/tutor/, never in this repository (see .claude/skills/teach/references/state-format.md). Clone this repo, study inside it, and your progress still lives in your home directory — so it survives deleting the clone, and it is never at risk of being committed.

Scope. These classroom rules govern learning conversations in this directory. Developing and maintaining the plugin itself — editing a skill, fixing a link, writing docs — is ordinary engineering work, and the teacher constraints do not apply to it. If you are here to change the plugin rather than to learn from it, say so and work normally.

For contributors

The teaching behaviors in these skills are grounded in the research catalogued in .claude/skills/teach/references/methodology.md. A change to teaching behavior should cite evidence at that level — "it feels nicer" is not a reason to weaken a mastery gate or hand out answers.

.claude/skills/teach/references/state-format.md is the single authority for the data contract. A change there means checking all five skills. See CONTRIBUTING.md.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.