Reader Entry Article Refiner
This is a publicly available entry skill.
It can be used independently, or as a post-processing node in the content workflow, specifically responsible for converting abstract judgments into entrances that readers are willing to click into.
Purpose
This is not a skill that "helps you come up with profound viewpoints".
This is a skill that "helps you write profound viewpoints in a way that people are willing to read".
It is specifically designed to solve the following problems:
- The viewpoint itself is correct, but it is too big, too distant, too abstract
- The opening talks about the essence right away, and readers have no sense of immersion
- The title looks like an author's summary, not a reader's confusion
- Golden quotes appear too early, making the article feel ungrounded from the beginning
- The article talks about "big propositions", but readers cannot feel what it has to do with themselves
The core task of this skill is only one:
Translate "what the author has already figured out" into "the sense of something wrong that readers have already felt".
When to use
When the content you write has the following characteristics, you must call this skill:
- A large number of abstract words appear in the title, such as:
- Subjectivity
- Right of judgment
- Cognitive structure
- Right of attention allocation
- Platform takeover
- System rewrite
- The article starts talking about "essence", "changes of the times", "deeper problems" right away
- You think the article is very good, but you vaguely feel that ordinary readers will not click it
- You find that your sentences "look like conclusions", not "like a sense of pain"
- You are writing about big topics such as AI, platforms, workflows, human value, etc.
Position in ark-skills
- Public role: Title and opening optimization
- Internal subskills:
subskills/pain-first-opening/SKILL.md
subskills/example-scene-selector/SKILL.md
subskills/anti-preach-rewrite/SKILL.md
subskills/distribution-hooks/SKILL.md
- Recommended collocation:
- First use to determine the main line
- Then use this skill to rewrite the main line into a reader entrance
- Can be used as an enhancement step after the first draft or optimized draft in
Internal Routing
When the problem is more specific, divert according to the following rules:
- The opening is too abstract, no sense of loss:
subskills/pain-first-opening/SKILL.md
- Examples only have task names, no story scenes:
subskills/example-scene-selector/SKILL.md
- The tone is like giving a lecture, like a methodology summary:
subskills/anti-preach-rewrite/SKILL.md
- The article is well written, but lacks hooks for likes, collections, and shares:
subskills/distribution-hooks/SKILL.md
Core principle
Always follow this order:
Symptom → Counterintuitive explanation → Mechanism → Essence → Landing point
Never write in this order:
Essence → Mechanism → Explanation → Example
The reason is very simple:
Readers do not click in for your "essential judgment".
Readers click in for their own "real-life confusion".
Non-negotiable rules
Rule 1: Conclusions cannot be used as the front door
The sharpest sentences should usually be placed in the middle and latter part of the article, as dimension-raising sentences or concluding sentences.
Do not use them directly as titles and openings.
Wrong example:
- What you lose is not attention, but the right to decide where to put your attention
Correct usage:
- Put the above sentence in the 60%–80% position of the article as a summary sentence
Rule 2: The title must write "symptoms" first, not "essence" first
The title prefers expressions like these:
- Why... instead...
- You obviously... why do you...
- You are not... you are just...
- When there are more and more..., why is it harder and harder for you to...
The title should make readers feel:
"This is the confusion I really have recently."
Instead of:
"The author of this article wants to discuss a very big problem."
Rule 3: Use reader language, not author language
Prioritize using these words:
- Read a lot
- Harder and harder to explain clearly
- More and more like repeating
- Feel like I know it
- But actually didn't figure it out
- Obviously more efficient, but more empty
Avoid using these words as the core of the title and opening:
- Subjectivity
- Right of judgment
- Cognitive structure
- Attention sovereignty
- Sorting right crisis
- Platform takeover
These words are not unusable.
They should be placed in the middle and latter part of the text to raise the dimension, not placed at the entrance.
Rule 4: An article can only have one entrance question
If you want to talk about:
- Attention
- Platform algorithm
- AI summary
- Subjectivity
- Human value
It means you haven't converged yet.
You must first compress it into a single entrance question, for example:
- The more AI helps you summarize, why is it harder for you to have your own opinions?
- Why do we read a lot of information every day, but have less and less judgment?
- You don't care less about the world, you just rely more and more on hot spots to decide what to care about
Only one can be selected.
Rule 5: Write "minimum perceptible loss" first, then write "big proposition"
Don't write directly:
- Human subjectivity is being taken over by platforms
Write first:
- Read a lot of summaries, but it's harder and harder to form your own opinions
- Scroll through a lot of information, but it's harder and harder to explain one thing clearly
- Obviously more efficient, but more and more like repeating others
Let readers feel the loss first, then explain the structure behind this loss.
Workflow
Step 1: Extract the "essence sentence" that the author really wants to say
Ask yourself first:
What is the one sentence I really want to say?
Examples:
- What you lose is not attention, but the right to decide where to put your attention
- AI improves efficiency, but may also weaken the intermediate process of people forming judgments
- workflow is not only amplifying you, it may also distill you
Keep this step first, don't rush to use it as a title.
Step 2: Find the "symptom sentences" that readers have already felt
Ask yourself again:
Where will readers really feel something wrong recently?
Write it into a life-like, sensory sentence.
Examples:
- Why do I read so much information, but have less and less judgment?
- Why the more AI summaries I read, the harder it is for me to have my own opinions?
- Why do I feel more and more like repeating, not thinking?
- Why is it obviously more efficient, but I feel more and more like a replaceable process?
Step 3: Do a "essence sentence → symptom sentence" translation
Translate abstract sentences into clickable reader questions.
For example:
Author's essence sentence:
What you lose is not attention, but the right to decide where to put your attention.
Translated entrance sentence:
Why do we read so much information every day, but have less and less judgment?
Another example:
Author's essence sentence:
AI has taken over the intermediate process of forming opinions for you.
Translated entrance sentence:
The more AI helps you summarize, why is it harder for you to have your own opinions?
Step 4: Verify if the title is "something readers would say themselves"
When checking the title, ask these three questions:
- Will ordinary readers regard this sentence as their own confusion?
- Does this sentence talk about symptoms first, not conclusions first?
- Can this sentence be understood within 3 seconds?
If any of the three items is no, keep modifying.
Step 5: Build the text with a fixed structure
The text must be expanded according to this structure:
1. Symptoms
Write about the real experience that readers already have.
2. Counterintuitive explanation
Tell him: The problem is not what you think it is.
3. Mechanism
Explain how platforms, AI, distribution, summarization, outsourcing work.
4. Essence
Then raise the article to a higher level, for example:
- Judgment formation mechanism
- Subjectivity
- Value being distilled
- Intermediate process disappearing
5. Landing point
Finally converge to a judgment that truly belongs to you.
Output format
Every time you run this skill, the output must include the following 5 parts:
1. Essence sentence
The core judgment that the author really wants to say, only keep 1 sentence.
2. Symptom sentences
The specific confusion that readers have already felt, write at least 3 alternative sentences.
3. Title candidates
Give 3 titles:
- The most stable one
- The most suitable for dissemination
- The most consistent with the author's positioning
4. Article skeleton
Strictly follow: Symptoms → Counterintuitive explanation → Mechanism → Essence → Landing point
5. Risk check
Point out the 3 most common mistakes this article is prone to make at present.
Anti-abstraction checklist
After writing, you must go through this checklist.
Title check
- Are there any big abstract words directly placed at the front of the title?
- Does the title look like "author's summary" instead of "reader's confusion"?
- Does the title clearly present a specific loss?
Opening check
- Does the opening talk about big principles right away?
- Is there any daily experience or obvious symptom in the opening?
- Can readers know "what problem of mine this article is talking about" within 5 sentences?
Text check
- Is the strongest golden quote placed too early?
- Are there more than two entrance questions stuffed in at the same time?
- Is there a situation of "raise dimension first, then land" leading to ungrounded content?
Ending check
- Is the ending just repeating the title?
- Does the ending raise the local symptoms to a higher-level mechanism?
- Does the ending leave a judgment that truly belongs to the author?
Example transform
Input
Author's essence sentence:
What you lose is not attention, but the right to decide where to put your attention.
Bad title
What the platform takes away is not your time, but your right to judge what is worth caring about
Problems:
- Too much like an author's summary
- Too abstract
- The cost is too distant
- Ordinary readers don't think this is their immediate problem
Better title
Why do we read so much information every day, but have less and less judgment?
Best title for AI context
The more AI helps you summarize, why is it harder for you to have your own opinions?
Style constraints
- Don't use "big words to oppress people" right at the beginning
- Don't pretend to be profound, let readers have a sense of immersion first
- Don't talk about all topics at once
- Don't put golden quotes at the title and opening
- An article is only allowed to have one entrance question
- All "essential judgments" must be landed through a specific symptom
Final command
When you think an article is "very good, but probably no one will click it", don't keep raising it.
Ask first:
In this article, what is the sense of something wrong that readers have already felt?
Write this out first.
The rest can be raised slowly later.