Verwendung
- Installationsbefehl kopieren und das Quellpaket vor der Installation in der eigenen Agentenumgebung prüfen.
- Benötigte Werkzeuge, Zugangsdaten und Hosts im Original prüfen.
- Eingabe und gewünschtes Ergebnis beschreiben und die Ausgabe mit den Quellen abgleichen.
npx skills add https://github.com/appeeky/ua-skills --skill competitor-ad-teardownVor dem Start
- Die Installation umfasst weder Modellabo noch API-Kontingent oder Dateizugriffsrechte.
- Sterne und Installationen zeigen Aktivität, keine unabhängige Qualitäts- oder Sicherheitsprüfung.
Originales Skill-Dokument
Aus dem Quell-Repository gerendert; Überschriften, Beispiele, Code, Tabellen, Links und Bilder bleiben erhalten.
Competitor Ad Teardown (Meta Ad Library)
Analyze competitor paid social creatives via Appeeky Meta Ad Library integration. Produce a swipe file of patterns and hooks — not a copy-paste kit.
Research before you generate. The best first batch mirrors proven formats with your unique positioning.
Initial Assessment
- Read
app-ads-context.md— your app, category, geo - Identify 1–3 competitors (direct + aspirational)
- Confirm read-only Ad Library access (Appeeky API key; optional
X-Meta-Ad-Library-Token)
What You Can Learn (and Cannot)
| Can learn | Cannot learn |
|---|---|
| Active creative formats | Exact spend / ROAS |
| Hook patterns and angles | Which ad "wins" internally |
| Long-running ads (proxy for profit) | Full funnel CVR |
| Platform mix (FB, IG, etc.) | Audience targeting details (limited) |
| CTA and copy structure | Proprietary assets to steal |
Long-running active ads (30+ days) are the strongest signal — competitors rarely keep losers live.
Data Sources (Appeeky MCP)
| Tool | Use when |
|---|---|
meta_app_ad_intelligence | You have competitor App Store / Play ID |
meta_ads_search | Keyword or brand name search |
meta_ads_page_ads | You know Facebook Page ID |
See appeeky-meta-ads.md.
Workflow
Step 1 — Identify Target
| Input | Example |
|---|---|
| Competitor app name | "Calm" |
| Apple App ID | 571800810 |
| Google package | com.competitor.app |
| Facebook Page ID | If known from prior search |
Pull organic context from aso-skills competitor-analysis if available — aligns ASO vs paid positioning.
Step 2 — Fetch Ad Intelligence (by app)
Best starting point — resolves app → brand → Ad Library:
meta_app_ad_intelligence
app_id: "571800810"
platform: "apple"
country: "us"
status: "ACTIVE"
media_type: "ALL"
limit: 25Google Play:
meta_app_ad_intelligence
app_id: "com.competitor.app"
platform: "google"
country: "us"
lang: "en"
limit: 25Override search terms if brand name differs from app title:
meta_app_ad_intelligence
app_id: "571800810"
platform: "apple"
query: "Calm meditation sleep"Step 3 — Search by Keyword or Page
Keyword search:
meta_ads_search
query: "Calm meditation"
countries: "US"
status: "ACTIVE"
media_type: "ALL"
limit: 25Page-specific:
meta_ads_page_ads
page_id: "<facebook-page-id>"
countries: "US"
status: "ACTIVE"
limit: 25Paginate with after cursor if limit exhausted.
Step 4 — Analyze Each Ad
For top 10–15 ads (prioritize long-running + diverse formats):
| Dimension | What to extract |
|---|---|
| Hook | First line of primary text / video opening |
| Format | Video / static / carousel / meme |
| Angle | Problem, social proof, offer, fear, aspiration |
| Visual | UGC, studio, screenshot, before/after |
| CTA | Install, Learn more, Sign up |
| Runtime | ad_delivery_start_time → today |
| Platforms | Facebook, Instagram, Messenger, Audience Network |
| Seasonality | Holiday, back-to-school, New Year |
Step 5 — Pattern Summary
Cluster ads into 3–5 patterns. Count frequency.
Step 6 — Translate to Your Variants
Map gaps to ad-creative-variants matrix — angles competitors don't use are opportunities.
Analysis Framework
Hook Taxonomy
| Type | Signal words | Example |
|---|---|---|
| Problem | "Tired of", "Still", "Can't" | "Still can't focus?" |
| Outcome | "Get", "Finally", "In X days" | "Sleep better in 7 nights" |
| Social proof | "Join", "million", "#1" | "Join 10M+ users" |
| Curiosity | "This", "Why", "Secret" | "This app fixed my mornings" |
| Identity | "For [persona]" | "For busy moms" |
| Offer | "Free", "Trial", "% off" | "7-day free trial" |
Format Taxonomy
| Format | Scroll-stop mechanism | Your Appeeky preset mapping |
|---|---|---|
| UGC phone | Authenticity | person_holding_phone + ugc |
| Split problem/solution | Contrast | problem_solution_split |
| Testimonial quote | Trust | creator_testimonial |
| Clean product shot | Premium | clean_app_store_mockup |
| Video demo | Motion proof | tiktok-creative-strategy (not static API) |
Longevity Scoring
| Days active | Interpretation |
|---|---|
| < 7 | Testing — weak signal |
| 7–30 | Promising — watch |
| 30–90 | Likely working |
| 90+ | Core evergreen creative |
REST API Equivalents
bash
# App intelligence
GET /v1/apps/:id/ad-intelligence?platform=apple&country=us
# Search
GET /v1/ads/meta/search?query=Calm&countries=US&status=ACTIVE
# Page ads
GET /v1/ads/meta/pages/:pageId/ads?countries=USHeaders: Authorization: Bearer <appeeky-key>; optional X-Meta-Ad-Library-Token.
Output Template
markdown
# Competitor Ad Teardown — [Competitor Name]
**App ID:** [id] | **Platform:** apple | **Ads analyzed:** N active
**Date:** [date] | **Geo filter:** US
## Executive Summary
[2–3 sentences: what they lean on, what's missing, your opportunity]
## Top Patterns
### 1. [Pattern Name] — X of N ads
- **Format:** Video / static / carousel
- **Angle:** Problem / proof / offer
- **Typical hook:** "..."
- **Why it works:** [hypothesis]
- **Example library ID:** [id]
### 2. [Pattern Name] — X of N ads
...
## Hook Swipe File
| Hook (paraphrased) | Type | Format | Days live | Platforms |
|--------------------|------|--------|-----------|-----------|
| | Problem | Video | 90+ | IG, FB |
| | Social proof | Static | 45 | FB |
## Visual Patterns
- Dominant: [UGC / studio / screenshot]
- On-image text: [heavy / minimal]
- Brand colors: [notes]
## Platform Mix
| Platform | % of ads |
|----------|----------|
| Instagram | |
| Facebook | |
| Audience Network | |
## Gaps & Opportunities
- **Angles they avoid:** [list]
- **Formats underused:** [e.g. before/after]
- **Weak CTAs to beat:** [generic "Learn more"]
- **Geo gaps:** [EU vs US creative differences]
## Recommended Variants for Us
→ Run `ad-creative-variants` with:
| Variant | Angle (inspired by, not copied) | Style / Preset |
|---------|--------------------------------|----------------|
| A | [your problem hook] | problem_solution / split |
| B | [your UGC angle] | ugc / person_holding_phone |
| C | [gap they miss] | professional / mockup |
## Ethical Use Reminder
- Inspire format and structure — never copy exact copy/visuals
- Do not present competitor ads as your ownMulti-Competitor Comparison
When analyzing 2–3 competitors:
| Pattern | Comp A | Comp B | Comp C | You |
|---|---|---|---|---|
| UGC video | ✅ Heavy | ⚠️ Some | ❌ None | ? |
| Before/after | ❌ | ✅ Heavy | ❌ | Opportunity |
| Free trial CTA | ✅ | ✅ | ⚠️ | Match |
TikTok Competitor Research (manual)
No TikTok Ad Library API in Appeeky yet:
- TikTok Creative Center → Top Ads by industry
- Search
#[competitor]and"[competitor] ad"on TikTok - Document in same teardown template under TikTok appendix
- Feed hooks into
tiktok-creative-strategy
Benchmarks (category signals)
| Signal | Saturated market | Open opportunity |
|---|---|---|
| Same hook in 80%+ ads | Parity required + differentiation | |
| Video dominance | Need video channel (TikTok/Meta video) | Static may still work if others ignore |
| No static ads | Static 1024 via meta-ad-creative | |
| Heavy offer ads | Trial/price in your copy | Brand/outcome angle unused |
Common Mistakes
| Mistake | Fix |
|---|---|
| Copying hooks verbatim | Paraphrase; adapt to your listing |
| Analyzing inactive ads only | Filter status: ACTIVE |
| One competitor only | Compare 2–3 for pattern confidence |
| Ignoring ad age | Weight long-running ads higher |
| Research without action | Always output variant recommendations |
| Skipping your own listing | Generate only after teardown → variants plan |
Ethical & Legal Use
- Inspire, don't infringe — formats and categories are fair game; assets and exact copy are not
- Paraphrase all hooks in swipe file
- Use competitor ads for internal strategy only
- When showing user examples, label clearly as competitor ad with source
Pre-Teardown Checklist
- [ ] Competitor app IDs confirmed
- [ ] Geo filter matches your launch market
- [ ] meta_app_ad_intelligence or search returned results
- [ ] 10+ ads analyzed (or documented if fewer exist)
- [ ] Patterns counted and ranked
- [ ] Variant recommendations mapped to ad-creative-variants
- [ ] Ethical use note included in outputRelated Skills
ad-creative-variants— build counter-creatives from teardownmeta-ad-creative— single creative from your listingtiktok-creative-strategy— video format selectionmeta-campaign-setup— deploy after creatives ready- aso-skills
competitor-analysis— organic/ASO side - aso-skills
competitor-tracking— ongoing monitoring app-ads-context— your positioning vs findings
