UGC
by
Creator decisions for ecommerce teams: score each piece of customer content on what it would earn as an ad, then license it, boost it, reuse it, or pass.
Customer content gets judged on how it performs organically, which is a measure of the creator's audience rather than of the asset. A video with forty thousand views from someone whose followers will never buy from you is worth less than a video with eight hundred views that shows the product solving the problem your ads are bad at explaining. Teams chase the first and ignore the second, then pay for licensing on the wrong thing. This skill scores each piece on what it would earn as an ad and returns license, boost, reuse, or pass.
What It Does
- Scores each asset on the attributes that predict paid performance, not organic reach.
- Reads the creator's audience against the customers who actually buy from you.
- Prices a licensing fee against the spend the asset could realistically support.
- Separates an asset worth boosting from one worth recutting yourself.
- Checks the claims in the content against what you are allowed to say.
- Returns license, boost, reuse, or pass on each piece, with the reason.
Use Cases
- Stop paying for licensing on the highest view count rather than the best asset.
- Find the low-reach video that would work as a scaling ad.
- Price a creator fee against what the asset can actually carry in spend.
- Catch an unverifiable claim before it runs as paid media.
Example Session
Which of this month's customer videos should we put money behind?
The best organic performer is the worst paid candidate.
- Asset | Organic views | Audience match | Paid fit score | Action
- Unboxing, 41k views | 41,000 | 12% | 3.1 | Pass
- Problem-solution, 800 views | 800 | 68% | 8.4 | License, $600
- Styling, 6.2k views | 6,200 | 54% | 7.1 | Boost as is
- Review, 2.1k views | 2,100 | 61% | 5.2 | Reuse the first 4 seconds
The unboxing reached forty-one thousand people who do not resemble your customers, and unboxings convert poorly in paid because the hook arrives too late. The eight hundred view video opens on the problem in the first two seconds and matches your buyers, so $600 for a license is cheap against the spend it can carry. The review is not usable whole and its opening is worth cutting into something else.
Why This Beats Prompting It Yourself
Organic Reach vs Paid Performance: The Hidden Risk
A generic prompt ranks by views, because views are the visible number and the obvious proxy for quality. Reach measures the creator's distribution, which you are not buying; you are buying the asset. Paying a licensing premium for the highest view count regularly buys the weakest ad in the set.
When You Still Need The Claim Checked
Customer content says whatever the customer believed, and the moment you put spend behind it that becomes your claim. This skill flags the sentence about results, timelines or comparisons that you cannot substantiate, before it runs rather than after a platform or a regulator raises it. That check is the one nobody remembers until it matters.
Known Limitations
Paid fit is a prediction from attributes that have correlated with performance in your account, not a guarantee, and a new format will score conservatively. Audience match needs creator audience data, which some platforms report only in ranges. Claim checking flags candidates for review rather than giving legal clearance. It sets the action and does not contact creators or launch ads.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/ugc/
SKILL.md
references/paid-fit.md
assets/
# one file per asset, with the score and reason
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from picking the best performing posts?
Organic performance measures the creator's audience. This scores the asset itself on the attributes that predict paid performance, checks the audience against your actual buyers, and prices a fee against the spend it can support. The ranking usually inverts.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
What does it need connected to work?
The assets with their organic metrics, your creative performance history, and your customer profile. Creator audience data improves the match score and the skill labels which pieces of it were unavailable.
Why recommend paying for a video with 800 views?
Because the price of a license tracks reach and the value of an asset tracks how it performs in paid. Those come apart constantly, and the gap is where the cheap wins are. $600 for something that can carry real spend is the best trade in that table.
Does it verify claims made by creators?
It flags them. A sentence about a timeline, a health outcome or a competitor comparison gets surfaced so a person can check it against what you can substantiate. That is a prompt for review, not clearance, and the distinction matters.
What is included with the skill?
The skill itself, the paid fit reference, the asset scoring format, and the claim flag list. It is free and security scanned. We re-publish the ZIP when the scoring changes, so download it again if your rankings start shifting.
Licensing a creator is a margin decision, not a vanity one.
Polar ties creator content to net revenue and contribution margin across channels, so whitelist, reuse or pass is a call you can defend.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from picking the best performing posts?","acceptedAnswer":{"@type":"Answer","text":"Organic performance measures the creator's audience. This scores the asset itself on the attributes that predict paid performance, checks the audience against your actual buyers, and prices a fee against the spend it can support. The ranking usually inverts."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"The assets with their organic metrics, your creative performance history, and your customer profile. Creator audience data improves the match score and the skill labels which pieces of it were unavailable."}},{"@type":"Question","name":"Why recommend paying for a video with 800 views?","acceptedAnswer":{"@type":"Answer","text":"Because the price of a license tracks reach and the value of an asset tracks how it performs in paid. Those come apart constantly, and the gap is where the cheap wins are. $600 for something that can carry real spend is the best trade in that table."}},{"@type":"Question","name":"Does it verify claims made by creators?","acceptedAnswer":{"@type":"Answer","text":"It flags them. A sentence about a timeline, a health outcome or a competitor comparison gets surfaced so a person can check it against what you can substantiate. That is a prompt for review, not clearance, and the distinction matters."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the paid fit reference, the asset scoring format, and the claim flag list. It is free and security scanned. We re-publish the ZIP when the scoring changes, so download it again if your rankings start shifting."}}]}</script>