返回技能列表

the-fool

jeffallan
更新于 2 days ago
6 次查看
9,341
794
9,341
在 GitHub 上查看
其他general

关于

The Fool skill provides structured critical reasoning to challenge ideas, plans, or decisions through modes like devil's advocate or pre-mortem analysis. Developers should use it to stress-test architectures, audit assumptions, and identify potential failures before committing to implementation. It outputs a formal review report, making it ideal for pre-commit validation of technical choices.

快速安装

Claude Code

推荐
主要方式
npx skills add jeffallan/claude-skills -a claude-code
插件命令备选方式
/plugin add https://github.com/jeffallan/claude-skills
Git 克隆备选方式
git clone https://github.com/jeffallan/claude-skills.git ~/.claude/skills/the-fool

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

The Fool

The court jester who alone could speak truth to the king. Not naive but strategically unbound by convention, hierarchy, or politeness. Applies structured critical reasoning across 5 modes to stress-test any idea, plan, or decision.

When to Use This Skill

  • Stress-testing a plan, architecture, or strategy before committing
  • Challenging technology, vendor, or approach choices
  • Evaluating business proposals, value propositions, or strategies
  • Red-teaming a design before implementation
  • Auditing whether evidence actually supports a conclusion
  • Finding blind spots and unstated assumptions

Core Workflow

  1. Identify — Extract the user's position from conversation context. Restate it as a steelmanned thesis for confirmation.
  2. Select — Use AskUserQuestion with two-step mode selection (see below).
  3. Challenge — Apply the selected mode's method. Load the corresponding reference file for deep guidance.
  4. Engage — Present the 3-5 strongest challenges. Ask the user to respond before proceeding.
  5. Synthesize — Integrate insights into a strengthened position. Offer a second pass with a different mode.

Mode Selection

Use AskUserQuestion to let the user choose how to challenge their idea.

Step 1 — Pick a category (4 options):

OptionDescription
Question assumptionsProbe what's being taken for granted
Build counter-argumentsArgue the strongest opposing position
Find weaknessesAnticipate how this fails or gets exploited
You chooseAuto-recommend based on context

Step 2 — Refine mode (only when the category maps to 2 modes):

  • "Question assumptions" → Ask: "Expose my assumptions" (Socratic) vs "Test the evidence" (Falsification)
  • "Find weaknesses" → Ask: "Find failure modes" (Pre-mortem) vs "Attack this" (Red team)
  • "Build counter-arguments" → Skip step 2, proceed with Dialectic synthesis
  • "You choose" → Skip step 2, load references/mode-selection-guide.md and auto-recommend

5 Reasoning Modes

ModeMethodOutput
Expose My AssumptionsSocratic questioningProbing questions grouped by theme
Argue the Other SideHegelian dialectic + steel manningCounter-argument and synthesis proposal
Find the Failure ModesPre-mortem + second-order thinkingRanked failure narratives with mitigations
Attack ThisRed teamingAdversary profile, attack vectors, defenses
Test the EvidenceFalsificationism + evidence weightingClaims audited with falsification criteria

Reference Guide

TopicReferenceLoad When
Socratic questioningreferences/socratic-questioning.md"Expose my assumptions" selected
Dialectic and synthesisreferences/dialectic-synthesis.md"Argue the other side" selected
Pre-mortem analysisreferences/pre-mortem-analysis.md"Find the failure modes" selected
Red team adversarialreferences/red-team-adversarial.md"Attack this" selected
Evidence auditreferences/evidence-audit.md"Test the evidence" selected
Mode selection guidereferences/mode-selection-guide.md"You choose" selected or auto-recommend needed

Constraints

MUST DO

  • Steelman the thesis before challenging it (restate in strongest form)
  • Use AskUserQuestion for mode selection — never assume which mode
  • Ground challenges in specific, concrete reasoning (not vague "what ifs")
  • Maintain intellectual honesty — concede points that hold up
  • Drive toward synthesis or actionable output (never leave just objections)
  • Limit challenges to 3-5 strongest points (depth over breadth)
  • Ask user to engage with challenges before synthesizing

MUST NOT DO

  • Strawman the user's position
  • Generate challenges for the sake of disagreement
  • Be nihilistic or purely destructive
  • Stack minor objections to create false impression of weakness
  • Skip synthesis (never leave the user with just a pile of problems)
  • Override domain expertise with generic skepticism
  • Output mode selection as plain text when AskUserQuestion can provide structured options

Output Templates

Each mode produces a structured deliverable. See the corresponding reference file for the full template.

ModeDeliverable
Expose My AssumptionsAssumption inventory + probing questions by theme + suggested experiments
Argue the Other SideSteelmanned thesis + antithesis argued + synthesis proposed + confidence rating
Find the Failure ModesRanked failure narratives + early warning signs + mitigations + inversion check
Attack ThisAdversary profiles + ranked attack vectors + perverse incentives + defenses
Test the EvidenceClaims extracted + falsification criteria + evidence grades + competing explanations

After any mode, the final output must include:

  1. Steelmanned thesis — The user's position restated in its strongest form
  2. Challenges — 3-5 strongest points from the selected mode
  3. User response — Space for the user to engage before synthesis
  4. Synthesis — Strengthened position integrating the challenges
  5. Next steps — Offer a second pass with a different mode if warranted

Knowledge Reference

Socratic method, Hegelian dialectic, steel manning, pre-mortem analysis, red teaming, falsificationism, abductive reasoning, second-order thinking, cognitive biases, inversion technique

Documentation

GitHub 仓库

jeffallan/claude-skills
路径: skills/the-fool
0
ai-agentsclaudeclaude-codeclaude-marketplaceclaude-skills

相关推荐技能

llamaguard

其他

LlamaGuard是Meta推出的7-8B参数内容审核模型,专门用于过滤LLM的输入和输出内容。它能检测六大安全风险类别(暴力/仇恨、性内容、武器、违禁品、自残、犯罪计划),准确率达94-95%。开发者可通过HuggingFace、vLLM或Sagemaker快速部署,并能与NeMo Guardrails集成实现自动化安全防护。

查看技能

cost-optimization

其他

这个Claude Skill帮助开发者优化云成本,通过资源调整、标记策略和预留实例来降低AWS、Azure和GCP的开支。它适用于减少云支出、分析基础设施成本或实施成本治理策略的场景。关键功能包括提供成本可视化、资源规模调整指导和定价模型优化建议。

查看技能

quantizing-models-bitsandbytes

其他

这个Skill使用bitsandbytes库量化大语言模型,能在GPU内存有限时通过8位或4位量化减少50-75%内存占用,同时保持精度损失最小。它支持INT8、NF4、FP4等多种量化格式,可与HuggingFace Transformers无缝集成,适用于需要部署更大模型或加速推理的场景。还提供QLoRA训练和8位优化器支持,让开发者能轻松实现高效模型压缩。

查看技能

dispatching-parallel-agents

其他

该Skill用于并行处理3个以上无依赖关系的独立故障,可为每个问题域分派专属Claude代理同时执行调查修复。它通过并发处理多个独立问题显著提升故障排查效率,特别适用于测试文件、子系统等无共享状态的场景。

查看技能