All skills
mattpocock avatar

/teach

@3216582 official
by Matt Pocockmattpocock/skills274k stars
22,991

Teach the user a new skill or concept, within this workspace.

Use this Skill: https://skilld.dev/gh/mattpocock/skills/teach

This session only. Nothing lands on disk.

LEARNING-RECORD-FORMAT.md

≈687 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Learning Record Format

Learning records live in ./learning-records/ and use sequential numbering: 0001-slug.md, 0002-slug.md, etc. Create the directory lazily: only when the first record is written.

They are the teaching equivalent of ADRs: they capture non-obvious lessons, key insights, and stated prior knowledge that will steer future sessions. They are used to calculate the zone of proximal development.

Template

# {Short title of what was learned or established}

{1-3 sentences: what was learned (or what prior knowledge was established), and why it matters for future sessions.}

That is the whole format. A learning record can be a single paragraph. The value is recording that this is now known and why it changes what to teach next, not in filling out sections.

Optional sections

Only include these when they add genuine value. Most records won't need them.

  • Status frontmatter (active | superseded by LR-NNNN): useful when an earlier understanding turns out to be wrong and is replaced.
  • Evidence: how the user demonstrated the understanding (a question answered, an exercise completed, prior experience cited). Useful when the claim might be revisited.
  • Implications: what this unlocks or rules out for future sessions. Worth recording when non-obvious.

Numbering

Scan ./learning-records/ for the highest existing number and increment by one.

When to write a learning record

Write one when any of these is true:

  1. The user demonstrated genuine understanding of something non-trivial: not just exposure, but evidence they can use the concept correctly. This sets a new floor for what to teach next.
  2. The user disclosed prior knowledge: "I already know X." Record it so future sessions don't re-teach it. Also record the depth claimed.
  3. A misconception was corrected: the user previously believed something wrong and now sees why. These are high-value: they predict future stumbling blocks for related topics.
  4. The mission shifted in response to learning: the user discovered they cared about something different than they thought. Cross-link to [[MISSION.md]] and update it.

What does not qualify

  • Material that was merely covered. Coverage is not learning. Wait for evidence.
  • Anything already captured tersely in [[GLOSSARY.md]] as a term definition. Don't duplicate.
  • Session-by-session activity logs. Learning records are not a journal: they are decision-grade insights.

Supersession

When a later record contradicts an earlier one (the user's understanding deepened or corrected), mark the old record Status: superseded by LR-NNNN rather than deleting it. The history of how understanding evolved is itself useful signal.

Source: SKILL.md on GitHub

No alerts1mo3 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    The 'teach' skill provides a framework for managing a learning workspace. It is classified as low risk primarily due to the potential for indirect prompt injection since the agent is instructed to summarize external content from resources provided in the workspace without explicit sanitization or boundary markers.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

Signed by skilld at 3216582. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated last month
disable-model-invocation
true
argument-hint
What would you like to learn about?
  • teaching
  • learning
  • curriculum
  • knowledge-management
  • interactive-lessons
  • spaced-repetition
  • skill-acquisition
  • zone-of-proximal-development

README badge

README badge for mattpocock/skills/teach

This skill sets up a stateful teaching workspace where an AI agent creates personalized lessons, reference materials, and learning records to teach a user a new topic over multiple sessions, grounded in the user's stated mission and zone of proximal development.

Generated from the current SKILL.md.

Does this skill create lessons automatically?
No. The skill is a framework for structuring teaching. You direct the AI agent on what to teach, and the agent generates lessons as HTML files, reference documents, and learning records within the workspace directory structure.
Can I use this to learn multiple topics?
Yes. Each topic uses its own workspace directory with its own MISSION.md, lessons, reference materials, and learning records. The agent tracks your progress across sessions using these files.
What happens if I don't have a clear learning goal?
The agent will ask you questions to establish a mission — the reason you want to learn the topic. This grounds all subsequent lessons in real-world goals rather than abstract concepts.
Does this skill require internet access?
The skill recommends finding high-quality external resources and links to them in lessons, but it does not require internet to function. All core teaching happens in local HTML files and markdown documents.
How does the agent know what to teach me next?
The agent reads your learning records and mission to calculate your zone of proximal development — the right level of challenge. It then designs the next lesson to target a gap in that zone.

Generated from the current SKILL.md. These answers refresh after source changes.