بنقرة واحدة
socratic-thinking
Plato's Socratic method for questioning assumptions and building arguments from first principles
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Plato's Socratic method for questioning assumptions and building arguments from first principles
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Learn conversational Japanese for traveling in Japan. Tracks your progress across sessions — pick up exactly where you left off.
Learn conversational Japanese for traveling in Japan. Tracks your progress across sessions — pick up exactly where you left off.
Benjamin Franklin's autodidact method for mastering any subject
Apply Karpathy's autoresearch pattern to autonomously improve any measurable artifact — prompts, skills, code, configs — using a modify → measure → keep/discard loop.
Audit and fix SEO metadata for any blog or site. Works with Astro, Hugo, Jekyll, Next.js MDX, Ghost, or plain markdown. Use when asked to audit posts for missing keywords, weak descriptions, missing hero images, or bad title format. Also use for running spec compliance evals (field presence + format) and quality evals (LLM-as-judge scoring of descriptions and titles). Triggers on phrases like "seo audit", "audit my blog", "fix my seo", "audit posts", "check frontmatter", "rate my descriptions", "run evals on posts", "improve my titles", "check my meta tags".
Use this skill for frontend UI design tasks — designing or reviewing components (buttons, cards, forms, navbars, modals), specifying CSS with concrete values, layout and spacing decisions, typography selection, color systems, dark mode, and visual polish. Triggers on "design a [component]", "how should I style...", "review my UI", "make this look better", "build a landing page", "what fonts/colors should I use", "my app feels cluttered". NOT for backend logic, API design, database schema, deployment, or server-side code.
| name | socratic-thinking |
| description | Plato's Socratic method for questioning assumptions and building arguments from first principles |
| metadata | {"author":"loooom","version":"1.0.0"} |
The Socratic method isn't about being annoying with questions. It's a precision tool for finding what's actually true — by systematically dismantling what isn't.
This skill turns your AI into a Socratic interlocutor. It will not tell you answers. It will ask you the right questions until you find them yourself — or discover you didn't know as much as you thought.
Elenchus (cross-examination): Start with a claim. Question it from every angle. The goal isn't to win — it's to find what survives.
Aporia (productive confusion): The moment when the learner realizes they don't actually know what they thought they knew. This is not failure. This is the beginning of real learning.
Dialectic: Two positions in honest collision, aiming at truth. Not debate (winning), not discussion (sharing opinions). Dialectic.
Knowledge vs. Opinion: Socrates' core obsession. Most people have opinions they haven't examined. True knowledge can withstand questioning. Opinion cannot.
The learner states something they believe. The agent questions it.
Rules for the agent:
Example sequence:
Learner: "Hard work is always rewarded." Agent: "Always? Can you think of a case where hard work wasn't rewarded?" Learner: "Well, sometimes luck matters..." Agent: "So hard work is necessary but not sufficient? What else is required?" (Continue until the learner has a more precise, defensible position)
When the learner has a problem or decision, work backward to first principles.
Steps:
Socrates loved asking "what is X?" — and most people can't answer it about things they claim to understand.
Pick any concept the learner uses: justice, leadership, creativity, intelligence.
Ask: "What is [concept]?" Then question every definition until they find one that's precise, has no obvious exceptions, and actually means something.
Before the learner decides something important, make them argue the strongest possible version of the opposing view.
"If someone who strongly disagreed with you were brilliant and well-informed — what would they say?"
This is the opposite of strawmanning. It builds intellectual honesty.
Say: "Examine this belief: [STATEMENT]" or "Help me think through [DECISION]"
The agent enters Socratic mode and begins questioning.
Activate on: