agentleFS
Sign inStart free
YujxZJCNYujxZJCN/teaching-skills/CLAUDE.md

teaching-skills

CLAUDE.md34 starsChanged 5 months agoMITView on GitHub

This repository is a Claude Code skill suite for university professors. When working in a session where these skills are installed: - Single-artifact requests go to the specific skill, not the pipeline: course-level design → course-designer · class-meeting materials → lesson-builder · graded instruments and rubrics → assessment-architect · anything about an individual student → student-mentor · checking submissions against a standard → submission-auditor · evidence analysis and career artifacts → teaching-reflector. - Extension skills: rendering actual slide decks/figures/posters →…

The file

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If a course_passport.yaml exists in the working directory, load it before asking the professor for context they already provided.

    line 26
  2. Very specific:

    No invented institutional policy, calendar dates, or learner context — ask or mark [NEEDS PROFESSOR INPUT].

    line 33
  3. Very specific:

    Never recommend AI-detection tools as an integrity mechanism (shared/ai_era_integrity.md).

    line 39

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 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

Teaching Skills — routing discipline

This repository is a Claude Code skill suite for university professors. When working in a session where these skills are installed:

Routing

  • Single-artifact requests go to the specific skill, not the pipeline: course-level design → course-designer · class-meeting materials → lesson-builder · graded instruments and rubrics → assessment-architect · anything about an individual student → student-mentor · checking submissions against a standard → submission-auditor · evidence analysis and career artifacts → teaching-reflector.
  • Extension skills: rendering actual slide decks/figures/posters → deck-studio · executable labs, datasets, starter code, autograders → lab-forge · video scripts, storyboards, captions → media-scripter · announcements, weekly emails, course site, LMS packaging → course-publisher · TA handbooks, grading calibration, allocation → ta-coordinator · accreditation matrices, evidence packages, self-study → accreditation-mapper · glossaries, translation, bilingual sync → bilingual-courseware · pre-lesson diagnostics, cohort readiness analysis, grouping (学情分析) → cohort-analyst (cohort aggregates only; an individual student's situation → student-mentor).
  • Lifecycle or multi-stage requests ("prepare my whole course", "new semester") → teaching-pipeline.
  • Ambiguous between guided and direct production → prefer the guided/Socratic mode; producing an unwanted full artifact wastes more than a clarifying dialogue.
  • If a course_passport.yaml exists in the working directory, load it before asking the professor for context they already provided.

Non-negotiables (all skills)

  1. Checkpoints per shared/checkpoint_protocol.md — the professor decides; surfaced defaults; "just proceed" respected.
  2. No invented institutional policy, calendar dates, or learner context — ask or mark [NEEDS PROFESSOR INPUT].
  3. Uncertain domain claims in generated teaching materials carry [VERIFY] markers.
  4. Person-affecting outputs (feedback, letters, interventions) are evidence-bound, draft-only, and end with a verify-before-send reminder. Student data never enters the Course Passport.
  5. Never recommend AI-detection tools as an integrity mechanism (shared/ai_era_integrity.md).

Editing this repo

  • A mode or agent change touches three places together: the skill's SKILL.md, MODE_REGISTRY.md, and docs/ARCHITECTURE.md.
  • New shared contracts go in shared/; skills reference them by path rather than duplicating content.
  • Slash commands in commands/ map 1:1 to registry entries.

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.

Check your setup, free