跳到主要内容
知仓学习社ZHICANG

context-switch-recovery

Reconstruct where you were and what's next after an interruption, so a broken focus doesn't cost you the whole thread. Use when asked where was I, I…

不碰外部(只输出文字)无严重或高危命中mohitagw15856/pm-claude-skills

它会碰到什么

扫了多少1 个文本文件,4 KB
它会碰到什么不碰外部(只输出文字)
命中总数0 处
命中统计严重 0 · 高 0 · 中 0 · 低 0

这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。

技能内容

Context-Switch Recovery

An interruption doesn't just cost the minute it took — it costs the ten minutes of re-immersion to find your place again, and for some brains the thread is just gone. This rebuilds it fast: from whatever you remember, it reconstructs what you'd done and what you were mid-thought on, hands you the single action to re-enter the task, and sets a breadcrumb so the next interruption costs less.

What This Skill Produces

  • The reconstructed state — what you'd already done, what you were in the middle of, and what was next, rebuilt from your fragments
  • The re-entry action — the single next move to get back in, not just oriented
  • The lost thread, recovered — the mid-thought or half-idea you were about to lose, captured
  • A breadcrumb for next time — a tiny habit (a one-line "I'm here, next is X" note) so future interruptions cost less

Required Inputs

Ask for these if not provided:

  • What you were doing — the task, as much as you recall
  • What you remember doing last — the last thing you completed or touched
  • Any half-thought — what you were about to do or think when interrupted
  • How long you've been away — a minute or since yesterday

Framework: Rebuild, Re-Enter, Breadcrumb

  1. Reconstruct from fragments. From whatever the person remembers, rebuild the task's state — done, in-progress, next — so the picture reforms.
  2. Recover the lost thread. Prompt for the half-formed thought or intention they were mid-way through, before it evaporates — that's the expensive part.
  3. Give one re-entry action. Not "resume the project" — the specific next move that drops them back into flow (reopen X, reread the last line, do the next tiny bit).
  4. Make re-entry easy. The action should require re-immersion, not re-deciding — lower the cost of getting back in.
  5. Set a breadcrumb. Teach the one-line "parking note" habit — before any interruption, jot "I'm here, next is X" — so the next recovery is instant.

Output Format

You were: [the task]

Reconstructed state: done [x] · mid-way on [y] · next was [z].

The thing you were about to lose: [the half-thought, captured].

👉 Re-enter with: [the single concrete action to get back in].

Breadcrumb for next time: before you get pulled away, drop one line — "I'm here, next is ___" — and recovery is instant.

Quality Checks

  • [ ] Rebuilds the task state from the person's fragments
  • [ ] Recovers the half-formed thought before it's lost
  • [ ] Gives one concrete re-entry action, not "resume"
  • [ ] The action aids re-immersion, not fresh decisions
  • [ ] Teaches the breadcrumb habit for next time

Anti-Patterns

  • "Just pick up where you left off" — that's the problem, not the fix.
  • Orienting without a re-entry action.
  • Missing the half-thought the person was about to lose.
  • A heavy note-taking system instead of a one-line breadcrumb.

Example Trigger Phrases

  • "I got interrupted and completely lost my place — where was I?"
  • "Help me pick back up on what I was doing before that call."
  • "I forgot what I was working on. Here's what I remember…"
  • "I was mid-thought and got pulled away — help me recover it."
  • "How do I get back into this after being distracted?"

想直接用这个技能?

本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。

同名技能的其他版本

有 3 个不同仓库或目录里都有叫 context-switch-recovery 的技能。它们内容并不相同,别混用: