RS-002 / MD
Download the framework
The six parts, the evidence labels, the adaptation rules, and the final checks.
RS-002 / Framework + Prompt + Template
Creator Style Clone
Study how a creator makes content people notice, watch, and act on.
Download or copy it
Start with the framework, then use the prompt or template.
Download the framework, copy the prompt, or grab the profile template. You don't have to give me your email for any of it.
RS-002 / MD
Download the research prompt
A reusable prompt for studying a creator's content without making up what you cannot see.
RS-002 / MD
Download the profile template
A place to record your sample, observations, patterns, open questions, and adaptation ideas.
RS-002 / MD
Download the Nate Herk example
A short example showing what the framework can pull out of the existing Nate Herk research.
Copy the research prompt Open prompt
That's the full prompt. Copy it here or download it above.
# Creator Style Clone Research Prompt
Version: 1.0.1
ResourceStak: RS-002 Creator Style Clone
## Use this prompt
Replace `[CREATOR]` with the creator you want to study. Replace `[TARGET CREATOR]` with your own name or project before you run the prompt.
```text
Study how [CREATOR] makes content.
I want to understand how this creator chooses ideas, packages them, builds each
piece, uses visuals, moves the audience to a next step, and gets the work made.
This is a study of content mechanics. It is not a request to imitate the
creator's voice, personality, biography, stories, worldview, catchphrases,
humor, appearance, or private material.
Use the creator's own website, videos, interviews, product pages, and public
accounts whenever you can. Use somebody else's summary to find leads, not as
proof. Build a sample before drawing conclusions. Ideally include 10 to 20
representative pieces, several strong performers, and several ordinary or
underperforming pieces. Record the date range, platform, titles, thumbnails,
transcripts or full text, descriptions, CTAs, linked resources, and any creator
interviews or behind-the-scenes material that you actually use.
Keep these labels separate:
- OBSERVED: I can point to it in the sample.
- REPORTED: the creator or team said it, but I didn't verify the internal claim.
- INFERENCE: this is my reading of what the evidence suggests.
- ADAPTATION: this is a mechanism [TARGET CREATOR] might change and use.
- UNKNOWN: the available evidence doesn't support the claim.
Don't invent a production team, software stack, hidden analytics, conversion
rate, reason a piece succeeded, or private workflow. Don't turn a price,
audience count, view count, or public result into a claim about income or
internal performance. If the evidence is thin, say UNKNOWN.
Study six parts:
1. WHAT THEY KEEP MAKING
- recurring topics and audience problems
- recurring promises and series
- evergreen versus timely ideas
- search versus recommendation intent
- beginner versus advanced subjects
- commercial intent and proof-of-work ideas
- strong, ordinary, and underperforming pieces
2. HOW THEY PACKAGE THE IDEA
- title patterns
- thumbnail patterns
- opening promise
- specificity and curiosity
- visible outcomes
- numbers and time framing
- beginner access
- first-person proof
- recurring shells
For every pattern, give examples, frequency in the sample, the job it might
do, conditions where it helps, conditions where it fails, and whether it
depends on the creator's identity.
3. HOW EACH PIECE UNFOLDS
- the first 15 to 60 seconds
- promise restatement
- proof timing
- problem setup and context
- teaching and story balance
- demonstrations and examples
- section order and pacing
- objections and recap
- ending and CTA timing
4. WHAT JOB THE VISUALS ARE DOING
- talking head, screen recording, live build, diagrams, slides, and B-roll
- charts, captions, overlays, zooms, and proof screenshots
- cut frequency and attention resets
For each recurring device, explain the job it seems to perform: prove,
explain, compress, re-engage, orient, build trust, or create pace. Label the
explanation INFERENCE when the creator didn't state it.
5. WHERE THE AUDIENCE GOES NEXT
- the primary next action
- CTA placement
- free resources, newsletters, communities, products, and services
- sponsors, affiliates, related content, descriptions, pinned comments,
and end screens
Give the viewer one clear next thing to do. A creator can have more than one
offer, but a single piece still needs a clear next move.
6. HOW THE WORK GETS MADE
- research, ideation, scripting, filming, editing, thumbnails, cadence,
repurposing, templates, automation, delegation, and quality control
Only include what the evidence supports. Keep private or undiscovered
details UNKNOWN.
Then produce:
1. A one-sentence style thesis.
2. A sample description with dates, platforms, pieces, strong performers, and
ordinary or underperforming performers.
3. An Idea Selection profile.
4. A Packaging pattern library.
5. A Content Architecture map.
6. A Visual Language map.
7. A Conversion Architecture map.
8. A map of how the work gets made.
9. A list of Style DNA that could be adapted.
10. A list of Identity elements that should not be copied.
11. Reusable content shells, with examples and limits.
12. A comparison of strong, ordinary, and underperforming pieces.
13. An adaptation map for [TARGET CREATOR]. For each adaptation, state the job
it performs and what [TARGET CREATOR] could do in their own way.
14. UNKNOWNs and open questions.
15. The complete source and sample list.
Finish with a short integrity check. Point out where the sample is thin, where
you made an inference, and where the evidence doesn't support a conclusion.
```
Read the output as research notes, not as a verdict. The labels are there to show what the evidence can carry.That's the full prompt. Copy it here or download it above.
What it is
A six-part way to study a creator's content without copying the person.
Study how a creator makes content people notice, watch, and act on.
Use it when
You might want this if you're...
- Creators who want to understand why somebody else's content works.
- Researchers who want to separate what they saw from what they're guessing.
- People who want to borrow a useful content mechanic without borrowing somebody else's identity.
What you get
The framework, the prompt, and a way to fill it in.
- Six-part framework for studying repeatable content mechanics.
- Research prompt for testing a creator and a real sample.
- Profile template for keeping observations, patterns, and questions together.
- Nate Herk example showing what you might adapt and what you should leave alone.
How to use it
Start with a real sample, then work through it.
- Pick the creator you want to study and write down what you're trying to understand.
- Build a sample you can actually inspect. Include ordinary work instead of picking only the hits.
- Write down what you can see before you explain why it might work.
- Work through the six parts in order: ideas, packaging, structure, visuals, next steps, and production.
- Separate patterns that repeat from choices that happened once.
- Finish with two lists: what you could adapt for your work, and what should stay with the original creator.
COMPACT EXAMPLE / NATE HERK
Here's what the framework picked up.
A compact example from the existing Nate Herk research. Download the full example for the detail.
Download the Nate Herk exampleAbout this resource
The useful details.
Public kit
GitHub
The public kit has the same framework, prompt, template, and example. It has its own clean history so you can clone it without getting Paul's private project history.
View the public kit on GitHubUsage and reuse
Use it, then adapt it.
Use this in your own work, inside a business, or with a client. Adapt the structure to fit what you're building. The public kit is released under CC BY 4.0, so you can use, share, and adapt it commercially when you follow the attribution and change-notice terms in its LICENSE file. Don't present Paul's original files as your own or use the framework to copy another person's identity, stories, voice, or private material. The license doesn't relicense third-party material.