메뉴
Trigger when users say "explain with a story," "tell me a fable," "give me an analogy," "ELI5," or "help me understand intuitively" about a concept. Uses fables to illuminate abstract concepts — the story never mentions the term, only revealing it at the end.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | concept-fable |
| description | Trigger when users say "explain with a story," "tell me a fable," "give me an analogy," "ELI5," or "help me understand intuitively" about a concept. Uses fables to illuminate abstract concepts — the story never mentions the term, only revealing it at the end. |
When a user wants to understand an abstract concept, don't throw a textbook definition at them. Instead, craft a carefully designed fable. The reader becomes immersed, only realizing near the end what it's about — then receives a clear explanation.
Core principle: Let them realize through story, not remember through definition.
Follow these 8 steps:
Relationship between core elements and metaphor points: Every core element must have a corresponding metaphor point that maps to it (this is the minimum). Additional metaphor points (3–5 total, including the ≥2 core mappings) enrich the story's details, but must not overshadow the core mapping. In short: Core elements: ≥2 must be mapped. Total metaphor points: 3–5.
Check with the user when:
Confirm understanding: "So this concept is essentially about ______, and the key insight is ______ — is that right?"
Confirm scene preference — assess the concept first, then decide whether to ask:
Story type — Choose based on the concept's core characteristics:
Selection criteria:
Narrative tone — Choose based on the concept's emotional flavor:
| Emotional Flavor | Recommended Tone | Example Scenario |
|---|---|---|
| Cautionary / warning (e.g., technical debt, deadlock) | Measured, slightly rueful | Campfire storyteller style, unhurried |
| Revealing / counterintuitive (e.g., leaky abstraction, survivorship bias) | Light, with a twist | Like telling a "guess what happened" story |
| Trade-off / dilemma (e.g., CAP theorem) | Neutral, impartial | Plain narration; let the reader decide |
| Mechanism / principle (e.g., dependency injection) | Everyday, down-to-earth | Casual, conversational — like explaining over coffee |
Simple concept exemption: For foundational concepts like "what is a variable" or "what is a loop," default to an everyday conversational tone — no need to match every row in the table above. The table is primarily for complex or abstract concepts that need emotional guidance.
Before writing a single word, verify that the chosen story type's conflict structure can faithfully reproduce the concept's causal chain from Step 1:
Three-act structure:
Writing principles:
Metaphor Concreteness Principle (hard requirement):
The metaphor vehicle must be something the reader can directly perceive in daily life. It must pass the Grandma Test:
Can your grandma understand, in one sentence, what the metaphor vehicle is? Yes → use it. No → swap it out.
---
## Explanation
### [Concept Name]
**One-sentence definition:** [Concise summary]
**Why this concept matters:** [1–2 sentences on practical significance]
**Key points:**
1. [What it is]
2. [Why it occurs / why it works this way]
3. [How to handle / apply it]
**Story-to-concept mapping:**
| Story Element | Concept Mapping |
|---------------|-----------------|
| [Character/Event] | [Concept element] |
| [Turning point] | [Core mechanism] |
**Prerequisite knowledge:** [If the explanation uses terms the reader might be unfamiliar with (e.g., using "bandwidth" to explain network latency, briefly explain bandwidth here in 1–2 sentences). Omit this line if not needed.]
**Food for thought:** [1–2 sentences to spark reflection]
**Story continuation (optional):** [If the concept has a naturally closely related concept (e.g., deadlock → deadlock prevention, npm → dependency hell), continue the story in 2–3 sentences to introduce the related concept. Give only a one-sentence definition, no full explanation template. End with "Want me to expand on this one?" Skip if no closely related concept exists.]
After completing the main explanation, proactively check whether a closely related concept exists (problem → solution, tool → common pitfall, phenomenon → adjacent phenomenon). If one does, don't skip the story continuation — it's one of the skill's most effective hooks.
Continuation constraints:
After writing the story, review the story against each item. Rewrite at most 2 times. If it still doesn't pass, output the current best version and honestly tell the user what you're not satisfied with:
concept-fable-{concept-name}.md.The following logic triggers after the user gives feedback on the output, not as part of the workflow steps:
The following logic fires before writing begins, to decide the story strategy:
These are common failure modes. Actively avoid them while writing:
| Anti-Pattern | What It Looks Like | Why It Fails | The Fix |
|---|---|---|---|
| Puppet characters | Characters do things against their own nature to serve the metaphor | The reader feels the characters are "unreal," and the story collapses | First conceive the character's plausible motivation, then find the coincidental resemblance to the concept — not the reverse |
| Over-metaphoring | Every detail maps to the concept; the story reads like a cipher | Becomes a textbook in disguise, loses all charm | Only map the core elements; let the remaining details serve the story's vividness |
| Premature reveal | "This is like concept X" said at the beginning or mid-story | Kills the suspense; the reader stops "realizing" on their own | Hold back — let the story reach its natural conclusion. The reveal belongs only at the end. |
| Hollow story | The story has no conflict, no tension | The reader won't remember it — no emotion, no memory | Ensure there's a clear dilemma, choice, or twist. Make the reader care about the characters. |
| ⚠️ Concept distortion | Twisting the concept's core meaning to make a better story | The most serious error — the reader learns an incorrect understanding | After writing, self-check: if a reader only read the story and not the explanation, would their understanding be correct? |
| ⚠️ Metaphor more abstract than concept | Using magic runes to explain leaky abstractions, or fantasy power-leveling systems to explain design patterns | The reader has to decode the metaphor itself first — the barrier isn't lowered, it's raised | Use the Grandma Test: swap the vehicle for everyday scenes (a universal remote, a translation app glitch) — ensure the vehicle itself requires no explanation |
If a concept genuinely doesn't suit the fable format (too abstract, no concrete equivalent), don't force it:
First, try to find the most concrete facet of the concept and write a story around that.
If there's truly no good story angle, switch to a "scene analogy" (a short metaphor rather than a full story). For example:
To explain "recursion": "Imagine you're standing between two mirrors. You see a reflection within a reflection within a reflection — each layer identical to the last, just smaller. But there's always a 'smallest you' at the deepest point. When light reaches there, the reflections start coming back layer by layer. That's recursion."
If the concept is too broad (e.g., "object-oriented programming"), suggest the user narrow it down (e.g., "encapsulation," "polymorphism").
If you can't find a scene the user is familiar with, be honest rather than forcing it. For example: "I'm struggling to find a scene you'd be familiar with for this concept — would it be okay if I used a more everyday scenario instead?" The user would rather accept a slightly less precise but natural story than sit through a forced, awkward analogy.
Each maps to a specific check in Step 7:
| Criterion | How to Verify |
|---|---|
| Story before concept | Self-check #3: Is the story readable on its own? |
| Subtle but not obscure | Self-check #2: Is the core tension precisely mapped? Does the causal chain align with the operating mechanism? |
| Accurate but not rigid | Self-check #1: Do the characters behave naturally? |
| Respect the reader's intelligence | Self-check #4: Is the reveal timed right? |
| One story, one concept | Self-check #5: Any redundant metaphors or plot points? |
| Metaphor lowers the barrier | Self-check #6: Does the metaphor vehicle pass the Grandma Test? |
| Explanation creates no new confusion | Self-check #7: Are all new terms briefly explained? |