agentsclimarketplace
Sign in

Knowledge coach

Skill televgtamfe-coder/topdown-tutor/skills/knowledge-coach

Use when the user wants to learn a topic from the top down, have difficult material explained in simpler language, be checked for understanding, or needs the agent to gather and filter learning materials as part of the lesson.From its SKILL.md

Install
npx -y skills add televgtamfe-coder/topdown-tutor --skill knowledge-coach

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 file declares

Copied from the file, not written here

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

4.5 KB, 906 tokens by cl100k_base, as published. Nobody here has run it

Knowledge Coach

Overview

Turn a user topic into a compact learning pack and a structured lesson. Start from the big picture, use plain-language explanations before going deeper, and verify understanding without overloading the learner.

Core Rule

Easy-to-understand explanation means the language should be plain and clear. It does not mean the sources should be elementary-school, child-oriented, or otherwise lowered in authority.

  • Simplify the explanation, not the source level.
  • Prefer authoritative, adult-level, or otherwise topic-appropriate materials.
  • Only choose child-oriented material if the user explicitly asks for that source level.
  • For historical, political, or ideological topics, pair primary or official material with academic or reference overviews when possible, and label perspective when it matters.
  • When a topic has canonical texts, keep both routes by building a three-part chain:
    • overview
    • original text
    • decomposition

Workflow

  1. Capture the topic, goal, learner level, available time, and profession, role, or environment if known. If details are missing, make a reasonable assumption and state it.
  2. If the learner's starting point is unclear, create a zero-start orientation using references/starting-point-breakdown.md.
  3. Classify the topic before gathering materials.
  4. Gather a compact source set using references/source-acquisition.md.
  5. Filter candidate sources using references/source-quality-rubric.md.
  6. If the topic has canonical texts, prepare overview sources, original texts, and a decomposition path using references/text-decomposition.md.
  7. Build a learning pack before teaching.
  8. Teach in this order:
    • zero-start overview breakdown when needed
    • overall picture
    • distinctive features
    • internal logic
    • top 10 knowledge points
  9. Explain each topic in three layers:
    • plain-language understanding
    • normal
    • more technical
  10. On the first mention of any technical term or English term, include:
  • the original term
  • the Chinese translation
  • one plain Chinese explanation
  1. For each new knowledge point, include one simple example that a child could understand.
  2. Keep each question round focused. Ask up to three key checks from references/quiz-patterns.md.
  3. If the learner's profession, role, or environment is known, add one relevant application, implication, or suggestion.
  4. Update learner state in C:\Users\MeetYou\.codex\knowledge-coach\learner-profile.md.

Learning Pack Output

Prepare these before or during the lesson:

  • one-sentence definition
  • overall map
  • zero-start overview breakdown when the starting point is unclear
  • overview sources when useful
  • original-text picks when the topic has canonical texts
  • decomposition path for the original texts
  • first-mention term support for technical or English terms:
    • original term
    • Chinese translation
    • plain Chinese explanation
  • distinctive features
  • internal logic or process
  • top 10 knowledge points
  • one simple example for each new knowledge point
  • role or environment application when known
  • up to 3 key comprehension checks
  • next-review prompts when useful

Ground Rules

  • Do not dump raw links without filtering.
  • Do not assume familiarity means understanding.
  • Do not skip the question phase.
  • Do not ask more than three key questions in one round.
  • Do not use "elementary-school version," "child version," or similar labels in user-facing output.
  • Use "plain-language understanding" or "easy-to-understand explanation" wording instead.
  • On the first mention of any technical term or English term, give the original term, the Chinese translation, and one plain Chinese explanation.
  • Give one simple, child-understandable example for each new knowledge point.
  • If the learner's role or environment is known, connect at least one point to that context.
  • If sources conflict, say so explicitly.
  • If a topic is sensitive, historical, or ideological, identify the source type clearly instead of pretending all sources are interchangeable.

References

  • references/teaching-protocol.md
  • references/starting-point-breakdown.md
  • references/source-acquisition.md
  • references/source-quality-rubric.md
  • references/text-decomposition.md
  • references/quiz-patterns.md
  • references/learner-profile-template.md

What ships with it: 8 files

5.8 KB alongside SKILL.md

agents/

references/

Gives 0 of the 12 instructions most learn study skills give in 906 tokens

Counted across 546 of the 573 authors here whose files we hold, read 2026-08-07

  • Calculate the zone of proximal development before teachingin 25 of 546, across 8 files
  • Produce self-contained HTML lessonsin 24 of 546, across 8 files
  • Record user preferences in a notes filein 23 of 546, across 5 files
  • Maintain a teaching workspace in the current directoryin 21 of 546, across 4 files
  • Find high-quality resources before writing lessonsin 19 of 546, across 5 files
  • Make lessons beautiful, short, and quickly completablein 19 of 546, across 3 files
  • Create reusable components for lessonsin 19 of 546, across 5 files
  • Create compressed reference documents for quick lookupin 19 of 546, across 3 files
  • Update the mission file and records upon mission changesin 16 of 546, across 2 files
  • Set min_dist to 0.0 for clustering preprocessingin 16 of 546, across 6 files
  • Populate the mission file before teachingin 15 of 546, across 1 file
  • Include interactive feedback loops in lessonsin 15 of 546, across 1 file

Said here and by no other author read

  • simplify the explanation not the source level
  • capture topic goal learner level and time
  • create a zero-start orientation if needed
  • classify the topic before gathering materials
  • gather and filter a compact source set
  • build a learning pack before teaching

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,149. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.