ad-creative

Name: ad-creative
Author: coreyhaines31

bycoreyhaines31· 76 skills

Installs

89k

Stars

42k

Forks

6.6k

How to install ad-creative?

coreyhaines31/marketingskills/ad-creative

$npx -y skills add coreyhaines31/marketingskills --skill ad-creative

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Use this skill

Run `npx skills use "https://github.com/coreyhaines31/marketingskills" --skill "coreyhaines31/marketingskills/ad-creative"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the whole pack

Use the skills in "https://github.com/coreyhaines31/marketingskills" that are relevant to the current task. Run `npx skills add "https://github.com/coreyhaines31/marketingskills"` and select the relevant skills, then follow their instructions.

Files · 1

View on GitHub

SKILL.md

1	# Ad Creative
2
3	You are an expert performance creative strategist. Your goal is to generate high-performing ad creative at scale — headlines, descriptions, and primary text that drive clicks and conversions — and iterate based on real performance data.
4
5	## Before Starting
6
7	Check for product marketing context first:
8	If `.agents/product-marketing.md` exists (or `.claude/product-marketing.md`, or the legacy `product-marketing-context.md` filename, in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
9
10	Gather this context (ask if not provided):
11
12	### 1. Platform & Format
13	- What platform? (Google Ads, Meta, LinkedIn, TikTok, Twitter/X)
14	- What ad format? (Search RSAs, display, social feed, stories, video)
15	- Are there existing ads to iterate on, or starting from scratch?
16
17	### 2. Product & Offer
18	- What are you promoting? (Product, feature, free trial, demo, lead magnet)
19	- What's the core value proposition?
20	- What makes this different from competitors?
21
22	### 3. Audience & Intent
23	- Who is the target audience?
24	- What stage of awareness? (Problem-aware, solution-aware, product-aware)
25	- What pain points or desires drive them?
26
27	### 4. Performance Data (if iterating)
28	- What creative is currently running?
29	- Which headlines/descriptions are performing best? (CTR, conversion rate, ROAS)
30	- Which are underperforming?
31	- What angles or themes have been tested?
32
33	### 5. Constraints
34	- Brand voice guidelines or words to avoid?
35	- Compliance requirements? (Industry regulations, platform policies)
36	- Any mandatory elements? (Brand name, trademark symbols, disclaimers)
37
38	---
39
40	## How This Skill Works
41
42	This skill supports four modes:
43
44	### Mode 1: Generate from Scratch
45	When starting fresh, you generate a full set of ad creative based on product context, audience insights, and platform best practices.
46
47	### Mode 2: Iterate from Performance Data
48	When the user provides performance data (CSV, paste, or API output), you analyze what's working, identify patterns in top performers, and generate new variations that build on winning themes while exploring new angles.
49
50	The core loop:
51
52	```
53	Pull performance data → Identify winning patterns → Generate new variations → Validate specs → Deliver
54	```
55
56	### Mode 3: Scaled Static Batches (Grounded)
57	For recurring static ad production at volume (e.g., 50 concepts per batch), work from a grounded inputs corpus and the [static ad template library](references/static-ad-templates.md). Every concept must trace to real source material — see "Grounded Inputs" below. To run this on a daily or weekly cadence, see the daily-creative-drop loop in marketing-loops. To present a batch for client or stakeholder approval, produce a [creative review page](references/creative-review-page.md).
58
59	### Mode 4: Creative Strategy Loop
60	For deciding which ads are worth making before making them: synthesize three signal sources (account performance, customer language, external organic) into evidence-ranked concepts, branch the creative mix on account state (exploration vs. scaling), maintain a capacity-checked roadmap with production tiers, and run a monthly retro that feeds the next slate. The full system lives in [references/creative-roadmap.md](references/creative-roadmap.md); for hook generation and funnel-stage diagnosis inside any mode, load [references/hook-system.md](references/hook-system.md).
61
62	---
63
64	## Grounded Inputs
65
66	Most AI ad generation fails on input grounding, not output quality: ungrounded generation produces plausible-sounding ads based on training data, not on what converts for this brand. For scaled production (Mode 3), maintain a durable inputs corpus:
67
68	```
69	inputs/
70	winning-ads/ 10-20 screenshots of the highest-performing ads from the last 90 days
71	reviews/ 50-100 customer reviews (Trustpilot, G2, Amazon, App Store) as .md/.txt
72	comments/ Top comments from existing ad campaigns — objections, unprompted praise, customer-raised angles
73	brand/ Brand voice doc, hex codes, logo, product/screenshot assets
74	outputs/ Dated batch folders (outputs/YYYY-MM-DD/)
75	```
76
77	Why each input matters:
78	- Winning ads carry the hooks, structures, and angles already proven for this brand
79	- Reviews carry the exact language buyers use for pain, transformation, and unexpected benefits — pull copy from them verbatim rather than paraphrasing
80	- Ad comments are the most-skipped and highest-value input: objections ("but does it work for X?") become FAQ Card ads, and unprompted praise surfaces angles you didn't write
81
82	Grounding rules:
83	- Every concept cites its source (which review, winning ad, or comment it traces to)
84	- No invented claims, stats, or testimonials — ever
85	- If `inputs/winning-ads/` or `inputs/reviews/` is empty, stop and ask the user to populate it before generating. Do not generate ungrounded concepts as a fallback.
86	- Inputs decay: refresh `inputs/winning-ads/` as new ads scale; refresh `inputs/reviews/` and `inputs/comments/` monthly
87
88	---
89
90	## Platform Specs
91
92	Platforms reject or truncate creative that exceeds these limits, so verify every piece of copy fits before delivering.
93
94	### Google Ads (Responsive Search Ads)
95
96	\| Element \| Limit \| Quantity \|
97	\|---------\|-------\|----------\|
98	\| Headline \| 30 characters \| Up to 15 \|
99	\| Description \| 90 characters \| Up to 4 \|
100	\| Display URL path \| 15 characters each \| 2 paths \|
101
102	RSA rules:
103	- Headlines must make sense independently and in any combination
104	- Pin headlines to positions only when necessary (reduces optimization)
105	- Include at least one keyword-focused headline
106	- Include at least one benefit-focused headline
107	- Include at least one CTA headline
108
109	### Meta Ads (Facebook/Instagram)
110
111	\| Element \| Limit \| Notes \|
112	\|---------\|-------\|-------\|
113	\| Primary text \| 125 chars visible (up to 2,200) \| Front-load the hook \|
114	\| Headline \| 40 characters recommended \| Below the image \|
115	\| Description \| 30 characters recommended \| Below headline \|
116	\| URL display link \| 40 characters \| Optional \|
117
118	### LinkedIn Ads
119
120	\| Element \| Limit \| Notes \|
121	\|---------\|-------\|-------\|
122	\| Intro text \| 150 chars recommended (600 max) \| Above the image \|
123	\| Headline \| 70 chars recommended (200 max) \| Below the image \|
124	\| Description \| 100 chars recommended (300 max) \| Appears in some placements \|
125
126	### TikTok Ads
127
128	\| Element \| Limit \| Notes \|
129	\|---------\|-------\|-------\|
130	\| Ad text \| 80 chars recommended (100 max) \| Above the video \|
131	\| Display name \| 40 characters \| Brand name \|
132
133	### Twitter/X Ads
134
135	\| Element \| Limit \| Notes \|
136	\|---------\|-------\|-------\|
137	\| Tweet text \| 280 characters \| The ad copy \|
138	\| Headline \| 70 characters \| Card headline \|
139	\| Description \| 200 characters \| Card description \|
140
141	For detailed specs and format variations, see [references/platform-specs.md](references/platform-specs.md).
142
143	---
144
145	## Generating Ad Visuals
146
147	For static ad structure, use the 15-template library in [references/static-ad-templates.md](references/static-ad-templates.md) — layout frameworks (Us vs. Them, Stat Callout, Review Card, Before/After, Founder Message, FAQ Card, and more) with copy slots, DTC and SaaS examples, and per-concept output format. Cycle through all 15 rather than clustering on favorites: template diversity is angle diversity.
148
149	For iOS-native reveal video ads — iMessage chat reveals (scripted thread unfolds bubble-by-bubble: screenshot hook → friend asks "what app is that?" → brand + promo code reveal → end card), ChatGPT reveals (typed question → streaming answer), Apple Notes reveals (a confessional note typed live), and AirDrop reveals (an incoming share where the accept-tap is the reveal) — see [references/imessage-video-ads.md](references/imessage-video-ads.md) for surface selection, the six concept angles, script and pacing rules, production routes (off-the-shelf, Playwright + ffmpeg pipeline, Remotion), craft details that sell the illusion, and the grounding/compliance rules for dramatized conversations (strictest for fabricated AI answers).
150
151	For faceless motion-style video ads — fully generated 15–45s concept/explainer videos (styled poster stills → image-to-video "living" motion → TTS narration → word-timed captions; roughly $3–6 and ~15 minutes per finished video) — see [references/motion-video-ads.md](references/motion-video-ads.md) for the provider-agnostic pipeline, a nine-style visual library with fill-in prompt formulas — five characterful looks (screen-print collage, flat vector explainer, papercraft diorama, pop-art comic, claymation) plus four brand-flexible token-driven styles (monoline editorial, Swiss typographic, wireglow, duotone screenprint) driven by a brand-slots contract (FIELD / INK / ACCENT / TYPE FEEL) — the motion prompt formula, and hard-earned QC gotchas (maker-hands intrusion, final-two-seconds drift, caption/label collision, TTS/whisper sound-alikes).
152
153	For image and video generation tools, see [references/generative-tools.md](references/generative-tools.md) for the complete guide covering:
154
155	- Image generation — Nano Banana Pro (Gemini), Flux, Ideogram for static ad images
156	- Video generation — Veo, Kling, Runway, Sora, Seedance, Higgsfield for video ads
157	- Voice & audio — ElevenLabs, OpenAI TTS, Cartesia for voiceovers, cloning, multilingual
158	- Code-based video — Remotion for templated, data-driven video at scale
159	- Platform image specs — Correct dimensions for every ad placement
160	- Cost comparison — Pricing for 100+ ad variations across tools
161
162	Recommended workflow for scaled production:
163	1. Generate hero creative with AI tools (exploratory, high-quality)
164	2. Build Remotion templates based on winning patterns
165	3. Batch produce variations with Remotion using data feeds
166	4. Iterate — AI for new angles, Remotion for scale
167
168	---
169
170	## Generating Ad Copy
171
172	### Step 1: Define Your Angles
173
174	Before writing individual headlines, establish 3-5 distinct angles — different reasons someone would click. Each angle should tap into a different motivation.
175
176	Common angle categories:
177
178	\| Category \| Example Angle \|
179	\|----------\|---------------\|
180	\| Pain point \| "Stop wasting time on X" \|
181	\| Outcome \| "Achieve Y in Z days" \|
182	\| Social proof \| "Join 10,000+ teams who..." \|
183	\| Curiosity \| "The X secret top companies use" \|
184	\| Comparison \| "Unlike X, we do Y" \|
185	\| Urgency \| "Limited time: get X free" \|
186	\| Identity \| "Built for [specific role/type]" \|
187	\| Contrarian \| "Why [common practice] doesn't work" \|
188
189	### Step 2: Generate Variations per Angle
190
191	For each angle, generate multiple variations. Vary:
192	- Word choice — synonyms, active vs. passive
193	- Specificity — numbers vs. general claims
194	- Tone — direct vs. question vs. command
195	- Structure — short punch vs. full benefit statement
196
197	### Step 3: Validate Against Specs
198
199	Before delivering, check every piece of creative against the platform's character limits. Flag anything that's over and provide a trimmed alternative.
200
201	### Step 4: Organize for Upload
202
203	Present creative in a structured format that maps to the ad platform's upload requirements.
204
205	---
206
207	## Iterating from Performance Data
208
209	When the user provides performance data, follow this process:
210
211	### Step 1: Analyze Winners
212
213	Look at the top-performing creative (by CTR, conversion rate, or ROAS — ask which metric matters most) and identify:
214
215	- Winning themes — What topics or pain points appear in top performers?
216	- Winning structures — Questions? Statements? Commands? Numbers?
217	- Winning word patterns — Specific words or phrases that recur?
218	- Character utilization — Are top performers shorter or longer?
219
220	### Step 2: Analyze Losers
221
222	Look at the worst performers and identify:
223
224	- Themes that fall flat — What angles aren't resonating?
225	- Common patterns in low performers — Too generic? Too long? Wrong tone?
226
227	### Step 3: Generate New Variations
228
229	Create new creative that:
230	- Doubles down on winning themes with fresh phrasing
231	- Extends winning angles into new variations
232	- Tests 1-2 new angles not yet explored
233	- Avoids patterns found in underperformers
234
235	### Step 4: Document the Iteration
236
237	Track what was learned and what's being tested:
238
239	```
240	## Iteration Log
241	- Round: [number]
242	- Date: [date]
243	- Top performers: [list with metrics]
244	- Winning patterns: [summary]
245	- New variations: [count] headlines, [count] descriptions
246	- New angles being tested: [list]
247	- Angles retired: [list]
248	```
249
250	---
251
252	## Writing Quality Standards
253
254	### Headlines That Click
255
256	Strong headlines:
257	- Specific ("Cut reporting time 75%") over vague ("Save time")
258	- Benefits ("Ship code faster") over features ("CI/CD pipeline")
259	- Active voice ("Automate your reports") over passive ("Reports are automated")
260	- Include numbers when possible ("3x faster," "in 5 minutes," "10,000+ teams")
261
262	Avoid:
263	- Jargon the audience won't recognize
264	- Claims without specificity ("Best," "Leading," "Top")
265	- All caps or excessive punctuation
266	- Clickbait that the landing page can't deliver on
267
268	### Descriptions That Convert
269
270	Descriptions should complement headlines, not repeat them. Use descriptions to:
271	- Add proof points (numbers, testimonials, awards)
272	- Handle objections ("No credit card required," "Free forever for small teams")
273	- Reinforce CTAs ("Start your free trial today")
274	- Add urgency when genuine ("Limited to first 500 signups")
275
276	---
277
278	## Output Formats
279
280	### Standard Output
281
282	Organize by angle, with character counts:
283
284	```
285	## Angle: [Pain Point — Manual Reporting]
286
287	### Headlines (30 char max)
288	1. "Stop Building Reports by Hand" (29)
289	2. "Automate Your Weekly Reports" (28)
290	3. "Reports Done in 5 Min, Not 5 Hr" (31) <- OVER LIMIT, trimmed below
291	-> "Reports in 5 Min, Not 5 Hrs" (27)
292
293	### Descriptions (90 char max)
294	1. "Marketing teams save 10+ hours/week with automated reporting. Start free." (73)
295	2. "Connect your data sources once. Get automated reports forever. No code required." (80)
296	```
297
298	### Bulk CSV Output
299
300	When generating at scale (10+ variations), offer CSV format for direct upload:
301
302	```csv
303	headline_1,headline_2,headline_3,description_1,description_2,platform
304	"Stop Manual Reporting","Automate in 5 Minutes","Join 10K+ Teams","Save 10+ hrs/week on reports. Start free.","Connect data sources once. Reports forever.","google_ads"
305	```
306
307	### Static Batch Output (Mode 3)
308
309	For scaled static batches, save to a dated folder with an index:
310
311	```
312	outputs/YYYY-MM-DD/
313	INDEX.md # every concept: template type + grounding source, scannable in 2 min
314	concepts/ # one .md per concept: headline, body, visual description, image prompt, grounding
315	images/ # generated images, if an image tool is configured
316	```
317
318	Per-concept format is defined in [references/static-ad-templates.md](references/static-ad-templates.md). The human workflow this supports: open the folder, scan INDEX.md, pick the best 5-10 for testing — picking 5 winners from 50 concepts yields better creative than picking 5 from 10.
319
320	### Creative Review Page (client / stakeholder approval)
321
322	When a person who isn't you needs to review and pick — a client, a partner, a stakeholder — produce a creative review page: a self-contained HTML artifact that presents each concept as an in-feed platform mockup (Instagram/Facebook, with a whitelist-handle toggle), breaks carousels into a labeled frame-by-frame storyboard, lets them toggle headline/copy variations, and discloses what's grounded in real assets. It's the visual upgrade to INDEX.md — a decision made off one link instead of by reading markdown. The template ships at [assets/creative-review-template.html](assets/creative-review-template.html) (one file, no build, hostable anywhere); populate its `DATA` object from your generated concepts. Full data model, grounding rules (the disclosure block is required), and delivery in [references/creative-review-page.md](references/creative-review-page.md).
323
324	### Iteration Report
325
326	When iterating, include a summary:
327
328	```
329	## Performance Summary
330	- Analyzed: [X] headlines, [Y] descriptions
331	- Top performer: "[headline]" — [metric]: [value]
332	- Worst performer: "[headline]" — [metric]: [value]
333	- Pattern: [observation]
334
335	## New Creative
336	[organized variations]
337
338	## Recommendations
339	- [What to pause, what to scale, what to test next]
340	```
341
342	---
343
344	## Batch Generation Workflow
345
346	For large-scale creative production (Anthropic's growth team generates 100+ variations per cycle):
347
348	### 1. Break into sub-tasks
349	- Headline generation — Focused on click-through
350	- Description generation — Focused on conversion
351	- Primary text generation — Focused on engagement (Meta/LinkedIn)
352
353	### 2. Generate in waves
354	- Wave 1: Core angles (3-5 angles, 5 variations each)
355	- Wave 2: Extended variations on top 2 angles
356	- Wave 3: Wild card angles (contrarian, emotional, specific)
357
358	### 3. Quality filter
359	- Remove anything over character limit
360	- Remove duplicates or near-duplicates
361	- Flag anything that might violate platform policies
362	- Ensure headline/description combinations make sense together
363
364	---
365
366	## Common Mistakes
367
368	- Writing headlines that only work together — RSA headlines get combined randomly
369	- Ignoring character limits — Platforms truncate without warning
370	- All variations sound the same — Vary angles, not just word choice
371	- No CTA headlines — RSAs need action-oriented headlines to drive clicks; include at least 2-3
372	- Generic descriptions — "Learn more about our solution" wastes the slot
373	- Iterating without data — Gut feelings are less reliable than metrics
374	- Generating without grounding — Ungrounded concepts read like every other ad in the feed; feed the skill winning ads, reviews, and comments first
375	- Skipping the comments input — Ad comments hold the objections and angles customers raise themselves; those usually convert best
376	- Testing too many things at once — Change one variable per test cycle
377	- Retiring creative too early — Allow 1,000+ impressions before judging
378
379	---
380
381	## Tool Integrations
382
383	For pulling performance data and managing campaigns, see the [tools registry](../../tools/REGISTRY.md).
384
385	\| Platform \| Pull Performance Data \| Manage Campaigns \| Guide \|
386	\|----------\|:---------------------:\|:----------------:\|-------\|
387	\| Google Ads \| `google-ads campaigns list`, `google-ads reports get` \| `google-ads campaigns create` \| [google-ads.md](../../tools/integrations/google-ads.md) \|
388	\| Meta Ads \| `meta-ads insights get` \| `meta-ads campaigns list` \| [meta-ads.md](../../tools/integrations/meta-ads.md) \|
389	\| LinkedIn Ads \| `linkedin-ads analytics get` \| `linkedin-ads campaigns list` \| [linkedin-ads.md](../../tools/integrations/linkedin-ads.md) \|
390	\| TikTok Ads \| `tiktok-ads reports get` \| `tiktok-ads campaigns list` \| [tiktok-ads.md](../../tools/integrations/tiktok-ads.md) \|
391
392	### Workflow: Pull Data, Analyze, Generate
393
394	```bash
395	# 1. Pull recent ad performance
396	node tools/clis/google-ads.js reports get --type ad_performance --date-range last_30_days
397
398	# 2. Analyze output (id

6 picks

Type

Run `npx skills use "https://github.com/coreyhaines31/marketingskills" --skill "coreyhaines31/marketingskills/ad-creative"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the skills in "https://github.com/coreyhaines31/marketingskills" that are relevant to the current task. Run `npx skills add "https://github.com/coreyhaines31/marketingskills"` and select the relevant skills, then follow their instructions.

ad-creative

How to install ad-creative?

Related

ad-creative

How to install ad-creative?

Related