القائمة
Use this before interpreting anything in the corpus. Attribute behavior correctly, then rebuild work episodes as the unit of analysis.
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/YannisKiefer/the-ai-diary --skill episode-methodيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
Use this to account for where AI-assisted attention actually went and what it bought - including the uncomfortable parts.
Use this as the always-on brain for any question about how someone thinks, decides, debugs, or works with AI. Read the request, find the first match, act.
Use this to test decision patterns against repeated behavioral evidence - without ever labeling from one anecdote.
عرض SKILL.md
| name | episode-method |
| description | Use this before interpreting anything in the corpus. Attribute behavior correctly, then rebuild work episodes as the unit of analysis. |
You are analyzing human + model interactions. Never attribute the model's behavior to the human.
Classify every observed action:
| Class | Meaning |
|---|---|
| USER INITIATED | The human asked, decided, or drove it |
| MODEL SUGGESTED | The model proposed it, human engaged |
| MODEL CAUSED | Model output produced the result |
| SYSTEM CONSTRAINT | Tooling, permissions, or environment forced it |
| UNKNOWN | Cannot tell - leave unknown, do not guess |
Worked examples of the trap:
Analyze within each model AND across models.
Individual messages are noise. The unit of analysis is the work episode.
An episode has up to ten stages: initial intention, problem framing, research, planning, execution, failures, corrections, verification, completion, later regression or reversal.
Follow tasks across multiple sessions where possible. The meaningful unit is DECISION -> ACTION -> RESULT -> UPDATE.
For each meaningful episode capture, when available: platform/model, timestamp, project, task type, intended outcome, complexity, initial framing, context supplied, success criteria set, planning performed, research performed, hypotheses considered, implementation started, major iterations, failures, corrections, model switches, tools used, validation, tests, shipped or not, later rework, later reversal, human correction required, repeated known mistake, lesson captured, lesson reused later.
Missing fields stay missing. Do not create false precision.
Attribute first. Episodes second. Messages last.