Borrowing it
Nothing to install: this file belongs to nukIeer/higgsfield-unlimited-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/nukIeer/higgsfield-unlimited-mcp/main/.claude/skills/viral-ugc-ads/SKILL.mdgit clone --depth 1 https://github.com/nukIeer/higgsfield-unlimited-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nukieer/higgsfield-unlimited-mcp/viral-ugc-ads)<a href="https://agentmods.dev/skills/nukieer/higgsfield-unlimited-mcp/viral-ugc-ads"><img src="https://agentmods.dev/badge/skills/nukieer/higgsfield-unlimited-mcp/viral-ugc-ads/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nukieer/higgsfield-unlimited-mcp/viral-ugc-ads"><img src="https://agentmods.dev/badge/skills/nukieer/higgsfield-unlimited-mcp/viral-ugc-ads.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00126 | $0.02459 |
| Opus 5 | $0.00063 | $0.01229 |
| Sonnet 5 | $0.00025 | $0.00492 |
| Haiku 4.5 | $0.00013 | $0.00246 |
Grade A, and why
viral-ugc-ads scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Viral UGC Ads (higgsfield-unlimited)
Build native-feeling, scroll-stopping short-form video ads. Native UGC feels specific, casual, and lightly imperfect — not a polished commercial. Describe creator behavior, not just the product.
0. Golden rules
- First 2 seconds decide everything. Lead every prompt with the hook — a problem, contradiction, curiosity gap, or hyper-specific moment.
- Make people slightly imperfect. Real humans have small flaws. Add them explicitly to every character/person prompt (see §4) or output looks fake.
- Use the best unlimited model, always. Unlimited is free — never settle for a weaker model to save cost. Pick by job (see §5).
- Test angles, change one variable at a time. Per concept generate ≥3 variations: problem-led, curiosity-led, benefit-led. Keep the rest constant.
- Research the GENERAL viral category, not only your niche. Copy the format of what's viral broadly (POV, "tell me without telling me", green-screen react, day-in-the-life, text-message story, satisfying loop) and pour your product in.
- Vertical 9:16, leave caption room (lower third + center), many watch muted.
- Claim safety. No medical/financial/body/unrealistic guarantees. Use experience wording ("here's what I noticed", "I like how…"). Never imply a fictional/AI creator is a real customer; no real celebrities or copied likeness.
1. The UGC prompt formula
Creator type + product + setting + first-2s hook + demo action + proof point
+ camera style + lighting + voiceover line + caption text + CTA + 9:16
Every field ties the output to a testable ad job. Be specific about the hook and the proof point ("my desk felt less cluttered" beats "amazing results").
2. The 12 copy-paste templates
Replace every [bracket]. Generate 3 hook variations of each.
- Talking-head review — "Create a vertical UGC talking-head of a [creator] reviewing [product] in [setting]. Hook: 'I didn't expect [benefit] to be this easy.' Handheld phone footage, natural indoor light, slight framing imperfections, captions <8 words, CTA: 'Try it if you want [benefit].' 9:16."
- Problem-solution — frustrated-with-[problem] shot → "This is what finally helped me with [problem]" → quick demo → soft result. Casual home light.
- Unboxing — hands opening package, texture close-up, reaction, one practical feature. Daylight, small camera shake, captions "First impression / Actually useful", CTA "Would you try this?"
- Beauty/skincare demo — mirror + soft morning light. Hook "My routine needed one simple upgrade." Texture on hand, natural application, calm reaction. NO before/after or medical claims.
- Fashion try-on — neck-down mirror, outfit reveal, 3 angles (front/side/ walk), fast jump cuts, captions "fit check / easy to style".
- App / SaaS walkthrough — "I found a faster way to [task]" → phone-screen shots of 3 simple steps. Show ONE task, not every feature.
- Gadget demo — "This tiny thing fixed my [problem]" → pick up/use → one clear result. Hands-on close-ups.
- Food / beverage — open on pour/bite/steam/plating. "I tried this because I wanted something [benefit]." Texture, serving, reaction. Warm kitchen light.
- Fitness product — "I needed something simple for [goal]." Use in home gym/outdoor, detail shots, realistic effort. No body-transformation claims.
- Faceless UGC — hands, product close-ups, phone screen, lifestyle only. Text overlay "I wish I found this earlier." Captions <8 words, CTA "See how it works." Great for product-image→video.
- Testimonial-style — "I've been using this for [time], here's what I noticed." Experience wording only; don't imply a real customer.
- Hook variation test — same demo, 3 openings: 1) problem 2) curiosity 3) result. Change ONLY the first line so you can judge the hook.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 170 lines · 126 tokens per session scan A 40ea752b882e
viral-ugc-ads is a skill published in the GitHub repository nukIeer/higgsfield-unlimited-mcp (3 stars, last pushed 1mo ago), licensed MIT. It adds 126 tokens to every session and 2,459 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…