How do I install the Learning Capture skill?

Run npx skillmds@latest add themarmack/learning-capture in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Learning Capture safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Learning Capture?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Learning Capture free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Learning Capture?

themarmack (@themarmack) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Learning Capture by themarmack — Agent Skill

Name: Learning Capture
Author: themarmack

Learning Capture

Save a structured insight note from an article, conversation, or experience. Routes through vault-writer to vault/insights/ (or vault/facts/{entity}/{predicate}.md if the captured content is atomic and falsifiable). Uses the controlled tag vocabulary. Distinct from quick-capture (which uses the qc-* sentence-starter formats and is faster). Use whenever encountering content worth retaining beyond the current conversation — a blog post that changes your mental model, a learning from a meeting that should outlast the meeting, a synthesis across multiple sources.

themarmack 1df361b 2.1 KB Updated Sep 17, 2026

File contents

learning-capture

The Category 6 PKM skill for "this is worth keeping." Distinct from quick-capture (which uses sentence-starter formats for fast turnaround) — learning-capture allows fuller framing when the input deserves it.

When to use

An article / talk / conversation changes a mental model.
A synthesis across multiple inputs is worth keeping as a single insight.
A "learnings from X" output that needs more structure than qc-insight provides.

When NOT to use

One-line capture → quick-capture (qc-insight format).
Multi-source research → Category 1 researcher skills.
Durable typed fact → write through vault-writer.write_fact directly.

Workflow

Input: source description + the learning itself + the trigger.
Classify: is it an insight (synthesis-shaped) or a fact (entity/predicate-shaped)?
Write via vault-writer.write_insight or write_fact accordingly.
Tag from the controlled vocabulary in _meta/tags.md.
Link to related vault notes via vault-querier for backlinks.

Output

vault/insights/YYYY-MM-DD-{slug}.md OR vault/facts/{entity}/{predicate}.md — always written via vault-writer.write_insight / write_fact (workflow step 3); the capture is never left only in chat.

Composes with

vault-writer — actual write.
vault-querier — suggest related notes for [[wikilinks]].
memory-curator — when uncertain whether durable.

Acceptance test

SKILL.md describes the classification logic. Live exercise deferred to next learning capture.

---name: learning-capturedescription: Save a structured insight note from an article, conversation, or experience. Routes through vault-writer to vault/insights/ (or vault/facts/{entity}/{predicate}.md if the captured content is atomic and falsifiable). Uses the controlled tag vocabulary. Distinct from quick-capture (which uses the qc-* sentence-starter formats and is faster). Use whenever encountering content worth retaining beyond the current conversation — a blog post that changes your mental model, a learning from a meeting that should outlast the meeting, a synthesis across multiple sources.---# learning-captureThe Category 6 PKM skill for "this is worth keeping." Distinct from `quick-capture` (which uses sentence-starter formats for fast turnaround) — `learning-capture` allows fuller framing when the input deserves it.## When to use- An article / talk / conversation changes a mental model.- A synthesis across multiple inputs is worth keeping as a single insight.- A "learnings from X" output that needs more structure than `qc-insight` provides.## When NOT to use- One-line capture → `quick-capture` (qc-insight format).- Multi-source research → Category 1 researcher skills.- Durable typed fact → write through `vault-writer.write_fact` directly.## Workflow1. **Input**: source description + the learning itself + the trigger.2. **Classify**: is it an insight (synthesis-shaped) or a fact (entity/predicate-shaped)?3. **Write** via `vault-writer.write_insight` or `write_fact` accordingly.4. **Tag** from the controlled vocabulary in `_meta/tags.md`.5. **Link** to related vault notes via `vault-querier` for backlinks.## Output`vault/insights/YYYY-MM-DD-{slug}.md` OR `vault/facts/{entity}/{predicate}.md` — always written via `vault-writer.write_insight` / `write_fact` (workflow step 3); the capture is never left only in chat.## Composes with- `vault-writer` — actual write.- `vault-querier` — suggest related notes for `[[wikilinks]]`.- `memory-curator` — when uncertain whether durable.## Acceptance testSKILL.md describes the classification logic. Live exercise deferred to next learning capture.

Frequently asked questions