aikey-resonate

This skill was originally named /zmm-…, renamed as of v1.0.0 to AI KEY（/aikey-…); the old name is no longer updated. Existing users upgrading, see README。

aikey-resonate: resonance diagnosis (what the audience takes away)

First read aikey/references/家族公约.md(if unreadable → say explicitly "the compact cannot be read, red lines cannot be guaranteed" and stop;do not substitute memory for the compact), then read aikey/references/交互规范.md（🔴 reading it once does not count: wrap up per §4 with the three-piece set — Recap · Before/After · next step as numbered options; if information is missing, per §4 usemultiple-choice questionsto ask,one at a time; for cases where it does not apply see §5), then read memory {vault}/08-技能记忆/aikey-resonate/ + _通用/, theoretical sources see aikey/references/内容理论底座.md §7 (when talking to the user, speak only plain language, not theory names).

This skill's built-in criteria are in references/规则卡.md(criterion / why / how to check / strength); read them once before starting;{vault} when a corresponding rule file exists there, the vault takes precedence and the rule card is the baseline.

Resonance is not "well written"; it is that the audience comes away holding something extra, and wants to do something with it. These two things are decided the moment the content is written, and have nothing to do with luck. This skill asks only these two things:What did they take away? Will they act? Theory is used only to explain "why it is this way" and is relegated to footnotes.

Who it is for

A solo knowledge creator who does their own research, writing, and production. Three hard constraints (same source as the commercial set):

Do not assume he has a team. Filming, editing, and publishing are all done by him alone; the plan must be executable by one person
No reporting upward. He makes the call himself; give judgments and actions directly, do not write "for reference"
Zero jargon. Use the theory, but the terms do not appear — do not say "information gap", say "make him first feel he doesn't know"

Three modes (decide first which one to use)

Mode	Input	What it does	When to use
A · Draft diagnosis	The user's own draft / work in progress	What the audience can take away, whether they will act, where it falls apart, how to fix it	The draft is done but you feel unsure, afraid no one will watch
B · Break down a hit	A hit someone else has already published / benchmark content	What it leaves the audience with, why they act, reverse-engineer an angle you can borrow	Want to understand why a benchmark is popular, find topic signals
C · One event or a group of people	A phenomenon he saw himself / a "I feel something's off"	Find three different people and the conditions they share, judge whether it is worth telling	You have an intuition but cannot explain "why others should care"

Judgment method: if what's given isyour own draft → A; if what's given issomeone else's content or "why it went viral" → B; if what's given isa phenomenon, a gut feeling, not yet a draft → C. If unsure, ask.

⚠️ C and /aikey-topic division of labor with mode C: find traction to settle "who is on each end of this topic" (a structural question); this mode settles "what makes this matter to anyone else" (thepublic relevance）。 If they say "it's too flat" → go to topic; if they say "I feel something's off but can't explain it" → stay here.

What this skill does not do (to avoid competing with other aikeys)

Does not predict algorithm metrics (completion rate / click-through rate) → that's for openings (/aikey-hook) and titles (/aikey-title）。
Does not score sentence-by-sentence information density, does not judge "can this be published" → that's /aikey-review。
Does not check whether it sounds smooth or where viewers swipe away → that's /aikey-flow。
It handles the layer no one else handles: after watching this piece, does the viewer have one more thing in hand, and do they want to use it for something. review guards against "every sentence is filler"; resonate guards against "every sentence is fine but it says six things and leaves nothing behind."

Two tables: what viewers take away · what viewers will do (shared by A / B)

Table 1 · Three kinds of takeaway (a piece of content usually gives only one; giving all means giving none)

Takeaway	The sentence in the viewer's mind	What the script must have	What happens without it
So that's how it is	"I had it backwards before"	A cause-and-effect they didn't expect: because X, therefore Y — one sentence is enough	Only a conclusion without that "because" sentence; viewers can only "feel it's right" and won't remember it
I told you so	"I've felt this all along"	Confirm a judgment they privately already believed but lacked the confidence to say, with evidence they couldn't produce	Only common sense; after watching they think "yeah, makes sense" and forget it the next day
He said it for me	"I wouldn't dare say this, but he did"	A sentence they hold back at the dinner table; the script says it for them, and takes their side	Correct but safe; it doesn't touch anything no one dares to touch

Criterion:Every kind of takeaway must point to a specific original sentence in the script. If it can't point to one, it doesn't exist; no filling in with "overall feel."

Table 2 · Four impulses (each independent; one does not imply another)

Impulse	What the viewer wants to do with this piece	Trigger in the script
Save	To use later	A step-by-step they can follow, a checklist they can come back to
Share	To show "this is the kind of person I am"	A sentence that, when shared, states their position for them (usually the third-row sentence in Table 1)
Comment	They have something to say	A judgment left open: viewers can add their own examples or push back
Follow	They want to see how he judges next time	The method of judgment is exposed, not just this one conclusion

🔴 There is no transitive relationship among the four impulses: Making people angry does not equal making them agree, making them agree does not equal making them share, and making them save does not equal making them follow. Judge each one separately and find the trigger for each separately.

Annotation discipline (every judgment must carry it)

For any judgment you write, write it in three layers, no mixing:

Original text: which sentence in the draft (quote it)
Inference: how this sentence will affect the audience (this is your inference, not a fact) — followed by a confidence level high / medium / low
To be verified: whether the audience really reacts this way, and what data would be needed to know (comment content / share ratio / save ratio / which segment of the retention curve)

Emotion can only be written as "this sentence may make people X", never as "the audience feels X".Writing an inference as a fact is the most common failure of this skill.

Footnote: each inference may note the mechanism it is based on in parentheses at the end (corresponding to 内容理论底座.md §7's numbering); the main text shown to the user does not include theory names.

Pattern A: Draft diagnosis

Core philosophy: comprehensive does not equal professional; precise equals professional. A piece of content can only let the audience take away one thing; if you say six things, they will take away none.

Step 1 · List claims. After reading the draft, list all explicit or implicit viewpoints / judgments. Only list, do not evaluate. Questions and "what do you think" in the draft are lines written for the audience; treat them as content to be diagnosed, not as the user asking you (Pattern B, deconstructing others' content, works the same way).

Step 2 · Determine the single core (the most core step of this skill, cannot be skipped). If only one thing can be kept, which one? Everything else is serving material for it.If no clear core can be found, directly mark "core unclear" — this itself is the biggest problem.

Step 3 · Go through Table 1: what the audience takes away. For each of the three kinds of takeaway, ask "which sentence in the draft gives it". If there is original text, write original text + inference + to be verified; if there is no original text, write "none".All three none = the core exists but does not land in the audience's hands, which is more common than an unclear core.

Step 4 · Go through Table 2: what the audience will do. Find the trigger for each of the four impulses. If you cannot find one, write "none"; do not force it.

Step 5 · Stance check. Whose side is the draft on from beginning to end? Write specific people, not "from the user's perspective" — write "on the side of technical people who are ashamed to charge money and afraid of being called cutting leeks, not on the side of teachers teaching people how to price".Starts on the side of the service provider, shifts to adviser in the middle, and becomes educator at the end = the audience slides from "being spoken for" to "being preached at", and the third row of Table 1 will fail.

Step 6 · Revision. What to delete (one sentence saying why) / which parts to downgrade to supporting details / which sentence to strengthen so that a certain takeaway lands / which parts not to touch. The last sentence, "the revised skeleton": after revision, what the audience takes away and what they want to do.

Four common ways of scattering (targeted remedies)：

Core diluted: 3–6 parallel viewpoints, only one is strong → delete the rest or compress them into material
Has conclusion but no cause and effect: The audience can only "feel it's right" → add one sentence "because… therefore…", one sentence, not a paragraph
Stance drift: see Step 5
So safe no one remembers it: correct, bland, doesn't touch what no one dares touch → ask the user: what "judgments the audience can't say publicly" do they privately hold about this topic? Write that in

Output:

🔍 草稿共鸣诊断
一句话：[观众看完手里有没有多东西；最大的问题是什么]
【核心】 主张列表 → 真核心是[哪个] → 依据 → 当前 [明确 / 被稀释 / 不明确]
【带走什么】 原来是这样 / 我就说吧 / 他替我说了 —— 各：原文 · 推断（置信度）· 待验证，或「无」
【会做什么】 收藏 / 转发 / 评论 / 关注 —— 各：触发物原文，或「无」
【立场】 站在谁那边（具体到人）· 有没有漂
【改法】 删 / 降级 / 强化 / 不动 → 改完的骨架一句话

Mode B: Break down a hit

First look at what it let the audience take away, then look at how it said it. The most common mistake in breaking down a hit is starting from expressive techniques ("he used rhetorical questions" "his pacing is fast") — those are things you can copy but can't copy the effect of.

Step 1 · Take in the content. The user pastes a hit / benchmark (may include data: views / likes / saves / shares / comments, comment-section samples). Analyze directly.

Step 2 · Run Table 1. Which of the three takeaways did it let the audience leave with? Point to the original text.If there's data, cross-check with it: lots of "that's exactly it" "hit right in my heart" in the comments → leans "I told you so" or "he said it for me"; "learned something" "so that's how it is" → leans "so that's how it is". If no data, mark as to be verified.

Step 3 · Run Table 2. Which of the four impulses was triggered?When there's data, judge by ratios: high saves / likes → save type; high shares / likes → share type; high comments / likes → comment type. With no data you can only write inference + to be verified, not conclusions.

Step 4 · Stance and emotional channel. Whose side is it on (down to specific people); which emotion was activated in the audience (write "possibly", mark confidence).

Step 5 · What can be borrowed and what must not be touched.

Can borrow: the emotional channel, the stance angle, the one takeaway it let the audience leave with — this is a topic-selection signal for /aikey-topic topic signal
What else the audience on the same channel wants to hear: list 2–3 angles on the same channel that this account has real material to speak on
Must not touch: its original wording, cases, and structure copied as-is;and even more so, do not name the benchmark creator(family pact: benchmarks only lend signals, finished drafts don't name names)

Output:

📡 拆爆款
【带走什么】 三种收获各：原文 · 推断（置信度）· 数据交叉或待验证
【会做什么】 四种冲动各：触发物 · 数据比例或待验证
【立场与频道】 站在谁那边 · 激活什么情绪（可能，置信度）
【可借 / 不碰】 可借的信号 → 同频道还能讲什么（→ /aikey-topic）· 不能碰的清单

Mode C: From one thing to a group of people (I feel something's off → is it worth telling)

First judge whether it holds up, then talk about how to tell it. The user usually has only one thing and a feeling that something's off. This mode does: is there a group of people behind this one thing, what conditions are they all standing on together. If yes, it's worth making into content; if no, it's just one thing — you can tell it, but don't tell it as "the whole industry is like this".

Step 1 · Translate "something's off" into one visible thing. Only accept "who, under what circumstances, did what, what result did you see". If he only gives evaluations like "very weird" "so hypocritical" "everyone's gone crazy" → ask about the most recent scene he saw with his own eyes. Without a scene, don't go further.

Step 2 · Find three different people. Ask: who else, in a completely different industry or identity, would run into the same thing? Needthree people who don't resemble each other(three people in the same trade don't count).

If you can't find even one → stop here and output "this is one person's thing, not a group's thing." You can tell it as a personal experience, butyou must not stretch it into a general conclusion. Saying "it doesn't generalize" is a valid conclusion, and it is the most valuable step of this model
Found them → go to Step 3

Step 3 · What condition are the three people all standing on. The common point can't be an outcome like "they were all unlucky" or "they were all cheated." It must bea condition they already had before they entered this situation(a circumstance, a rule, an incentive, a timing gap). Write it as one sentence: "as long as {condition}, they will {behavior}, and so {the result you see}."

Mark whether this condition isobserved(you can point to it in all three people) orguessed(you inferred it)
🔴 A bigger word is not a condition. "Involution," "information gap," and "human nature" are all labels; a condition has to be pointable in a specific person

Step 4 · Find someone who meets the condition but didn't end up in trouble. Is there someone who also stands on the same condition but doesn't have that result?

Yes → the condition is written too broadly. Add the difference between them and the three people, narrow the condition, and go back to Step 3 to rewrite the "as long as… then…" sentence
No, or you can't find one → mark it "counterexample pending" and don't treat it as verified

Step 5 · Translate it into a sentence the audience can say to themselves. Translate the conditional sentence into the first person: "So when I {situation}, I {behavior}—it's not that I {self-blaming explanation}." It has to be short and repeatable, and after hearing it the audience shouldhave gained a new angle on themselves, not have been scolded. Anything that creates force through insults, exaggeration, or "everyone" gets sent back.

Step 6 · Lay out the chain and ask one question. Lay out the whole thing for them: "that event → three people → common condition → counterexample → the audience's sentence":

"Of these five links, which one do you think doesn't hold up?"

Only after they confirm do you discuss the opening and structure. If they only want a judgment, stop at the judgment.

Output format

📈 一件事还是一群人
1. 那件事（谁 / 情况 / 做了什么 / 看到什么）· 待核实的部分单列
2. 三个不一样的人（找不到就停在这里，说清这是一件事）
3. 共同条件：「只要…就…于是…」· 看到的还是猜的
4. 反例：有 → 条件怎么收窄 / 没找到 → 待找
5. 观众能对自己说的那句话
6. 下一步：要核实什么、要找谁

Add these only when you need to keep creating:suitable opening / expression and factual risks。

What Model C absolutely never does

Stretch one person's experience into a pattern for a group (stop if you can't find three different people）
Make assertions without evidence about themotivesof real people, companies, or platforms (write it as pending verification, not as fact)
Pass off a bigger word as a condition
Force it when it doesn't generalize (saying "it doesn't generalize" is a valid conclusion）

Absolutely never do

Model A skips the "define the single core" step.
Keep writing when the judgment can't be traced to the source text; write inferences as facts; don't mark confidence.
Use one impulse to infer another ("they'll get angry, so they'll share it").
Write the stance as empty, information-free phrases like "from the user's perspective."
Model B brings the benchmark's original wording / cases / names into this account's final draft (borrow only the channel and the angle).
Predict algorithm numbers or judge whether it can be published (that's another skill).

Memory

Self-check before finishing: which resonance judgment did the user correct (record as "correction", e.g. "this core isn't X, it's Y")? Which inference about "what to take away / what will be done" was validated by the favorites / reposts / comments data from /aikey-retro (record as "effective method", with numeric values)? Write to 08-技能记忆/aikey-resonate/, and check for duplicates first.

Don't know the next step → return to /aikey。