Learned

gasserane/personal-skills/skills/learned

作者 gasseranea22368a1505356d9815ee7aa7ffd65f0afeb5bad無授權條款收錄於 2026年10月9日更新於 2026年10月9日

Fast post-deliverable learning capture. Use when Ane types "/learned", or says "capture what I learned", "log a learning", "what did I learn from this", or wants a quick 4-line note after finishing a piece of work. Appends a dated entry to the running learning log in the Obsidian vault. The explicit-command sibling of journal-reflection's Post-deliverable capture mode; for end-of-day, weekly, decision-review, or new-framework reflection, use journal-reflection instead.

AI 產生的概覽

在完成交付項目後,以四個問題快速擷取學習重點,並附加到 Obsidian 學習日誌中。

功能
透過四個單行問題進行簡短的交付後反思:這項工作教會了什麼、下次會怎麼做、要延續的一點、以及哪些做法值得重複。它會把四個答案以帶日期的條目附加到 Obsidian 儲存庫中的單一學習日誌檔案;若檔案不存在,則先建立含 frontmatter 的檔案。若無法存取儲存庫路徑,則在回覆中傳回 markdown 並說明目標路徑。它也可能提議將可重複的系統行為記錄為 feedback 記憶。
適用情境
適用於完成一項實質工作後,想快速記錄經驗教訓時。它是交付後擷取模式的顯式指令入口;更完整的日終、每週、決策回顧或新框架反思應改用 journal-reflection 技能。
執行需求
需要位於設定路徑(OBSIDIAN_VAULT_ROOT)的 Obsidian 儲存庫以寫入檔案;否則以 markdown 形式傳回筆記。不包含指令碼、套件或憑證。

/learned — post-deliverable learning capture

A 30-second human-learning loop: capture what a finished piece of work taught you, before it evaporates. This is the explicit-command entry point to the Post-deliverable capture mode defined in the journal-reflection skill — that skill holds the canonical procedure; this one is the fast /learned trigger. Origin: video-insights improvement #14.

When to use

After finishing any substantive deliverable, when you want to bank the lesson. For a fuller reflection (end-of-day, end-of-week, a past decision, or a new framework), use journal-reflection instead — those are its other four modes.

/ wrap-up already offers this at session end; /learned is for any other moment.

What to do

Ask these four, one at a time. Keep each answer to one line. Do not auto-answer; Ane's words go in verbatim.

  1. What did this deliverable teach you about the work?
  2. What would you do differently next time?
  3. One thing to carry forward.
  4. What worked that you want repeated? (success extraction)

If Ane does not name the deliverable, ask for a 2–4 word label.

Success routing. If answer 4 names a repeatable system behaviour (a prompt pattern, a workflow, a specialist or model choice), offer to also bank it as a feedback memory with Why: and How to apply: lines — one yes/no, no push. This gives /improve-system confirmed successes to trend, not only corrections. The vault entry records all four answers either way.

File placement

Append to the single running log 5 JURNAL/Learning/deliverable-learning-log.md in the Obsidian vault (OBSIDIAN_VAULT_ROOT = C:/Users/AGasser/OneDrive/Ane Obsidian Vault), exactly as the journal-reflection Post-deliverable capture mode specifies:

  • If the file does not exist, create it with YAML frontmatter (type: learning-log) and an H1, then add the first entry.
  • If it exists, append a new ## YYYY-MM-DD — <deliverable label> section with the four answers as bullets, preserving all prior entries byte-identical (edit-preservation; apply mel_wiki/wiki/concepts/edit-preservation-protocol.md).

Local-only: if the vault path is not reachable (web session, other device), return the markdown in the reply and name the target path so Ane can save it manually.

Limitations

Do not invent reflections. Do not lecture. One capture per invocation. This skill does not replace journal-reflection — it is the quick post-deliverable slice of it.

來源與署名

來源:gasserane/personal-skills位於skills/learned提交a22368a

授權條款: 無授權條款

內容歸原作者所有。SourceWeft 從公開儲存庫中收錄這些內容。

檢舉或申請下架