From logseq
ALWAYS invoke when the user says "log this learning", "log this as a learning", "I just learned", "I learned that", "remember this learning", "add to learnings", "note this as a learning", "learning:", or invokes /learn. Write to both today's journal and the permanent learnings page.
How this skill is triggered — by the user, by Claude, or both
Slash command
/logseq:learnThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Capture something learned — goes to today's journal AND the permanent learnings page.
Capture something learned — goes to today's journal AND the permanent learnings page.
0. Read graph root
grep '^graph_root=' ~/.logseq-plugin-config 2>/dev/null | cut -d= -f2
If this returns empty, tell the user to run /setup and stop.
Use the returned path as $GRAPH_ROOT for all file paths in this skill.
1. Extract the learning
Everything after the trigger phrase. E.g.:
"I just learned that useOne appends id even when empty" → the full discovery"log this learning — Prettier useTabs adds alignment spaces ESLint rejects" → after the dash"/learn React strict mode double-invokes effects in dev" → everything after /learnIf nothing follows the trigger, ask: "What did you learn?"
2. Classify into a category
| Category | When to use |
|---|---|
| Frontend | React, TypeScript, Vite, CSS, browser behaviour, component patterns |
| Backend & API | API contracts, Refine hooks, REST behaviour, database |
| Tooling & DX | git, ESLint, Prettier, Vitest, build tools, Claude Code |
| Architecture & Patterns | design decisions, abstractions, file structure |
| Debugging | how a bug was found, non-obvious root cause, debugging technique |
| Other | anything else |
3. Append to today's journal under ## Learnings
date "+%Y_%m_%d"
date "+%H:%M"
In $GRAPH_ROOT/journals/YYYY_MM_DD.md, append:
- ## Learnings
- `HH:MM` [learning] #learning #[category-tag]
logged:: [[YYYY-MM-DD]]
category:: [Category name]
Category tags: #frontend #backend #tooling #architecture #debugging
4. Append to pages/learnings.md under the matching section
Find the section header (e.g. ## Frontend) and append:
- [learning]
logged:: [[YYYY-MM-DD]]
ref:: [[YYYY_MM_DD]]
If there's useful context (file, hook name, API, what the fix was), add it nested:
- [learning]
logged:: [[YYYY-MM-DD]]
ref:: [[YYYY_MM_DD]]
- Context: [one sentence — file, hook, API, etc.]
Skip the nested context bullet if the learning is self-contained.
5. Confirm in one sentence
"Logged in today's learnings and saved to [Category] in the permanent page."
logged:: uses dashes (2026-05-13); ref:: uses underscores in double brackets ([[2026_05_13]])learnings.md is the permanent, searchable record; the journal is the dated contextnpx claudepluginhub kaushalkishormishra/logseq-plugins --plugin logseqGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.