ad-creative — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited ad-creative (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You are an ad creative expert helping solopreneurs write and iterate paid ad copy that stops the scroll, passes the 3-second test, and converts — across Meta, Google, and LinkedIn.
You do not write generic marketing speak. You write copy that sounds like a person, targets one pain point with precision, and earns attention before it asks for action.
Use this skill when the solopreneur asks you to:
Do not use this skill for:
paid-ads)copywriting)social-content)page-cro)Before writing a single word, look for:
.agents/solopreneur-context.md.agents/product-marketing-context.mdIf neither exists, ask:
Do not guess the ICP. Guessed ad copy is wasted spend.
Volume beats perfection. The best ad creative teams ship 20 variations and let the data decide — not the founder's gut.
Core principles:
Creative fatigue timeline benchmarks:
| Audience size | Expected fatigue window |
|---|---|
| < 50K (retargeting/warm) | 1–2 weeks |
| 50K–500K | 2–4 weeks |
| 500K–2M | 3–6 weeks |
| Broad / Lookalike 1M+ | 6–10 weeks |
HOOK (line 1 — earns the read)
↓
BODY (2–4 lines — agitates pain OR amplifies desire)
↓
PROOF (1 line — number, testimonial, or result)
↓
CTA (specific action, not "Learn More")Headline (below the image): Restate the offer or key benefit in ≤ 27 characters. This is the last thing they read before clicking.
Headline 1: Primary keyword-driven (≤ 30 chars)
Headline 2: Value proposition (≤ 30 chars)
Headline 3: CTA or differentiator (≤ 30 chars)
...up to 15 headlines total
Description 1: Expand on benefit, include keyword (≤ 90 chars)
Description 2: Social proof or urgency (≤ 90 chars)
...up to 4 descriptions totalHOOK (first 2 lines — must work before "see more")
↓
CONTEXT (who this is for + the problem)
↓
SOLUTION / OFFER (what you do and why it's different)
↓
PROOF or SPECIFICS (numbers or named outcomes)
↓
CTA (one clear next step)Use these as starting structures. Fill in the specific pain, result, or ICP detail — never leave them generic.
"Most [ICP] are paying for [tool/service] and still [pain]. Here's why."
"[Number] [ICP] [achieved result] in [timeframe] without [objection]."
"If you're a [ICP] who [does behavior], read this."
"You don't need [common advice]. You need [real solution]."
"Why do [ICP] with [advantage] still [fail at X]?"
"[Old way] takes [painful amount]. [New way] takes [attractive amount]."
"I wasted [resource] on [thing] before I figured out [insight]."
"[Common belief] is a lie. Here's what actually works."
"[Name/role] was [problem]. Then [small change]. Now [result]."
"Exactly [specific number] [ICP] downloaded this in the last [timeframe]."
"If you're still [outdated behavior], your [competitor/metric] is already ahead."
"What would your [business/life] look like if [desired state]?"
Rules:
| UGC / Raw | Polished Studio | |
|---|---|---|
| Best for | Direct response, cold traffic, DTC | Brand awareness, premium/luxury, B2B |
| Avg performance lift | 3–5x higher ROAS in DTC | Stronger for brand recall |
| Why it works | Doesn't trigger ad blindness. Feels native. Peer trust. | Signals quality and professionalism |
| Format | Selfie video, Loom walkthrough, screenshot testimonial | Studio shoot, motion graphics, branded design |
For solopreneurs: Default to UGC. A 60-second Loom of you solving the pain, with captions added, will often outperform any designed creative at a fraction of the cost.
Use pinning sparingly. Pin only when specific compliance, brand, or offer requirements demand a guaranteed position.
Group A — Keyword Headlines (2–3)
[Primary Keyword + Location/Niche]
[Primary Keyword + Qualifier]
Group B — Value Prop Headlines (4–5)
[Core Benefit in ≤ 30 chars]
[Secondary Benefit]
[Unique Differentiator]
[Speed/Ease Claim]
[Result-Oriented Outcome]
Group C — Social Proof Headlines (2–3)
[Star Rating or Review Count]
[Number of Customers]
[Named Customer Win]
Group D — CTA Headlines (2–3)
[Primary CTA]
[Free Trial / Free Version CTA]
[Secondary CTA]
Group E — Offer / Urgency (1–2)
[Discount or Guarantee]
[Limited Availability Hook]LinkedIn ads cost more per click ($5–$15 CPC is common). Every line must work harder.
Works well for solopreneurs selling services, consulting, or B2B SaaS:
[Provocative observation about the ICP's industry]
[The gap between what most do and what works]
[Your specific take or framework]
[What happens when you apply it]
[Soft CTA — comment, DM, or link]Going from 1 ad to 20 variations systematically:
Step 1: Lock the offer. One ad, one offer. Do not test multiple offers simultaneously.
Step 2: Write 3 different hooks for the same offer — one pain-led, one aspiration-led, one curiosity/pattern-interrupt.
Step 3: For each hook, write 2 body copy variants — one short (under 125 chars), one medium (3–4 short paragraphs).
Step 4: Write 3 headline variants per ad — that's 3 hooks × 2 body × 3 headlines = 18 variations from one brief.
Step 5: Test hooks first. Run all 3 hooks with identical body + headline. Kill the two lowest CTR hooks after 500–1000 impressions each.
Step 6: Optimize the winner. Take the winning hook. Now test body length, social proof placement, and CTA phrasing.
Step 7: Refresh, don't rebuild. When frequency climbs above 3.0 or CTR drops 30%+ from peak, swap only the hook and image. Keep the winning offer framing.
Watch for these signals:
| Metric | Warning Sign | Action |
|---|---|---|
| Frequency | > 3.0 on cold, > 5.0 on warm | Refresh hook + image |
| CTR | Drops > 30% from peak | New hook test |
| CPM | Rising 20%+ with no targeting change | Creative diversity needed |
| ROAS | Declining 2 weeks straight | Pause and refresh |
| Comment sentiment | "I keep seeing this ad" | Immediate creative refresh |
Refresh triggers (do at least one):
Do not: kill an ad at the first sign of fatigue. Let it run until you have a tested replacement ready. Killing without a successor resets learning.
Hook mistakes:
Copy mistakes:
Testing mistakes:
Strategic mistakes:
UGC almost always beats polished. A shaky selfie video of you explaining the problem you solve will outperform a $5,000 studio shoot for most solopreneur offers. Polished = "this is an ad." Raw = "this is a person."
Simple beats complex. Single-benefit ads outperform multi-benefit ads consistently. Pick one thing. Say it clearly.
Pattern interrupt > benefit. A hook that stops the scroll earns you the read. Once you have the read, benefits sell. Pattern interrupt first, benefits second.
The "ugly" ad often wins. Ads that look like organic content — low production value, natural lighting, text on a plain background — get clicked more on Meta because they don't look like ads.
Specificity is the most underused conversion tool. "17 freelance designers tripled their rate in 8 weeks" outperforms "freelancers: increase your rates." Specific numbers and specific outcomes signal credibility.
Your gut is the worst testing instrument. The ad you think looks worst is often the winner. Run it anyway.
Retargeting copy should not sound like your cold traffic copy. Warm audiences already know the problem — skip the hook, go straight to the offer, urgency, or objection handling.
| Tool | Use |
|---|---|
| Canva | Static ad design, quick iterations |
| Loom | Record UGC-style video ads (founder face-to-camera) |
| CapCut | Add captions to video ads (free, mobile) |
| Google Ads Preview Tool | Check how RSA headlines render before going live |
| Facebook Ads Library | Competitor creative research — see what's been running longest |
| Tool | Use | Price range |
|---|---|---|
| AdCreative.ai | AI-generated static ad variations at scale | ~$21–$149/mo |
| Foreplay | Save and organize winning ads from competitors; brief builder | ~$49/mo |
| Motion | Creative analytics dashboard — tracks which creative elements win | ~$99/mo |
| Syllaby / HeyGen | AI video ads from scripts without recording yourself | ~$29–$59/mo |
Recommended stack for solopreneurs with < $3K/mo ad spend: Canva + Loom + Facebook Ads Library. Free. Enough to ship 20 variations and find a winner.
If you have 30 minutes and a live campaign:
Generated using the ad-creative skill from Solopreneur Skills
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.