Cornell notes

Skill andresnator/agents-orchestrator/skills/cornell-notes

Install

npx -y skills add andresnator/agents-orchestrator --skill cornell-notes

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Trigger: cornell note, cornell notes, nota cornell, learning lesson capture. Markdown Cornell note format for micro-lessons: cue questions + notes + learner-voiced summary; cues feed quizzes and the spaced-repetition queue.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.4 KB, 527 tokens by cl100k_base, as published. Nobody here has run it

Cornell Notes

Activation Contract

Use when a learning flow captures a micro-lesson as a note: the 10% formal step of the learning-loop, or any time the user asks for a Cornell-style note on a concept just studied. Notes live at .ai/learning/<topic-slug>/notes/NNNN-<lesson>.md, numbered sequentially from 0001.

Do not use for book-chapter synthesis — that is the summarize skill; this skill defines the capture format inside a learning path.

Hard Rules

Follow assets/cornell-template.md: metadata header, Mermaid map, cue/notes table, summary, recall hand-off.
Cues are retrieval questions, never topic labels: "What does the filter chain decide per request?" not "Filter chain". Each cue must be answerable from its Notes cell alone.
3–7 cues per note; notes cells stay concise (2–4 lines each). If a lesson needs more, it is two lessons.
The Summary is the learner's voice: ask them to state it in 2–3 sentences and record what they say (lightly cleaned up); never invent it. If they can't, that is a signal — reteach before closing.
Every note embeds at least one Mermaid diagram (mindmap for concept overviews, graph TD for processes, sequenceDiagram for interactions) and cites at least one primary source in the header.
Every cue is handed to spaced-recall as a new card, and the note's Recall hand-off line lists the card IDs. Cues are also the topic's quiz bank.
Notes are Markdown in English; never HTML.

Format

See assets/cornell-template.md. Structure:

# NNNN — {Lesson title} + header quote block (topic, module, date, sources).
## Map — Mermaid diagram of the lesson's concepts.
## Notes (Cornell) — two-column table | Cue (question) | Notes |.
## Summary — 2–3 sentences in the learner's own words.
## Recall hand-off — card IDs added to review-queue.md.

Output Contract

Return the note path, the list of new cues (as questions), and confirmation that each cue was scheduled via spaced-recall. Flag any cue the learner could not answer during capture so the loop can reinforce it.