菜单
Sub-orchestrator for diagnostic questions. Routes to UAUA exploration, causal tracing, safety analysis, or direct debugging based on how much is known and the nature of the problem.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | diagnose - Find the Cause |
| description | Sub-orchestrator for diagnostic questions. Routes to UAUA exploration, causal tracing, safety analysis, or direct debugging based on how much is known and the nature of the problem. |
| output | {"format":"prose"} |
Input: $ARGUMENTS
What is observably wrong? State it in concrete terms. "It's broken" is not a symptom — what specifically is broken? What happened vs what was expected?
If the input is too vague, ask: "What specifically is happening that shouldn't be?"
Verify the problem is actually complex before applying complex analysis. Sometimes the obvious cause is the right one.
Before deep diagnosis, check:
| Situation | Mode |
|---|---|
| User wants quick answer | → /ezy (easy mode) |
| User wants maximum rigor | → /hrd (hard mode) or /certainty |
| User wants general principle | → /genl (what general pattern explains this?) |
| User wants specific application | → /spcf (apply known diagnosis framework to this case) |
| Problem seems too complex | → /dcm (decompose first) |
| User might be wrong about the problem | → /sid (situation identification) |
| User might have wrong mental model | → /rmm (recover from wrong mental model) |
Default path (cause unknown, no timeline): → INVOKE: /uaua [symptom] — explore possible causes, then test each
With timeline: → INVOKE: /fowwr [symptom + timeline]
With suspected cause: → INVOKE: /araw [suspected cause]
Recurring problem: → INVOKE: /sbfow [symptom]
| Situation | Also invoke |
|---|---|
| Diagnosis involves ethical dimensions | → /eth |
| Worst case matters | → /dys (dystopia analysis) |
| Need to trace implications of the cause | → /sycs (so you can see) |
| Need to check what else might break | → /ata (and then also) |
| Unresolved sub-questions in diagnosis | → /tbd (to be determined) |
| Self-deception about the cause | → /sdc (self-deception check) |
| Need the story of what happened | → /story (narrative construction) |
| Need to differentiate between similar causes | → /difr (differentiation reasoning) |
Report:
After diagnosis is complete, the user may need:
Route any input through a branching question tree to narrow down the optimal response strategy before writing. Two stages — PERCEIVE (classify input) then ACT (select response). Covers all prompt types.
Generate exhaustive guesses about user input using ALL search methods with coverage tracking. Guessing is SEARCH through possibility space. Tracks space created vs space covered to ensure comprehensive exploration.
Systematically evaluate and select from a set of guesses, options, or possibilities. Combines ARAW analysis with prioritization to determine which guesses are strong, weak, actionable, or eliminable.
Extract hidden assumptions from any content. Surfaces what must be true for claims to hold, enabling deeper analysis.
Sub-orchestrator for analytical requests. Routes to decomposition, systems analysis, comparison, risk assessment, or synthesis based on what kind of analysis is needed.
Assume Right - Deep recursive rightness search. For every claim, assume it's right — find what must follow, then assume THOSE implications are right too. Recurse until bedrock. Track every claim found.
