design.md ×

package-lock.json ×

package.json ×

plan.md

# anekaria.com — Product Plan

> **AI Variety Show & Entertainment Format Studio.** A creative studio that uses AI to generate, prototype, and produce entertainment formats — game shows, sketch series, talk show concepts, reality formats — for OTT platforms, broadcasters, and creator networks. Name = *aneka* (ID: variety) + studio energy.

## 1. The wedge — why anekaria wins
- **No category leader for "AI-native format development."** Studios like Endemol Shine, Banijay, and FremantleMedia license formats but development is slow (12–18 months, $200K+ per format bible). Local production houses in SEA can't afford that pipeline. We compress it to weeks for one-tenth the cost.
- **AI is finally good enough for format ideation.** GPT-5 + Claude 4.6 can produce defensible format treatments, pilot scripts, and visual mood boards in hours that previously took a writer's room weeks. Plus video pre-viz via Sora 2 / Veo 3.
- **Indonesian + SEA-first wedge.** OTT platforms (Vidio, Viu, Vision+, Disney+ Hotstar, IDN Media) commission 200+ locally-relevant formats per year. None have an AI-native development partner. Win this niche → expand to MENA + LATAM.
- **Production-ready output, not just decks.** Every format ships with bible + 3-episode pilot scripts + visual treatment + budget + casting brief — assembled by AI, polished by human producers. The "studio" not the "tool" positioning.

## 2. Money stack (3 tiers, USD priced)

| Tier | Price | Output | AI cost | Margin |
|---|---|---|---|---|
| **Format Spark** | $499 one-time | 1 format treatment (5pg) + mood board + 1 pilot scene | ~$30 | 94% |
| **Format Bible** | $4,900 one-time | Full format bible (40pg) + 3 pilot scripts + visual treatment + casting brief + budget framework | ~$300 | 94% |
| **Studio Partner** | $24,000/quarter | 6 format bibles per quarter + monthly review + production support + format licensing rep | ~$2,400 quarterly | 90% |

**LTV proxy:** Studio Partner repeat rate = 75% (production houses develop continuously). Format Bible upsells to Studio Partner = 38%.

**Year-1 target:** $400K (8 Studio Partners + 20 Format Bibles + 30 Format Sparks).

## 3. Acquisition (triple-stack)
1. **LinkedIn long-form B2B** — weekly essay aimed at heads of development at SEA OTT platforms. Hook: "Why your network commissions the same 4 formats every year." Build authority → inbound qualified leads.
2. **Showreel via TikTok/Reels** — 60-sec format pitches done in AI animation, shown publicly. Free creative samples that demonstrate quality. 1 reel per week.
3. **Direct outreach to development heads at Vidio, Viu, Vision+, IDN, BRIO, GoPlay** — handwritten-style intro email + a custom 3-format teaser already built for their slate gap.

## 4. AI automation map
| Loop | Model | Cost guard |
|---|---|---|
| Format ideation (concept generator) | Claude Sonnet 4.6 via OR | $0.08/format |
| Format bible drafting (40-page) | Claude Opus 4.7 via OR | $1.20/bible |
| Pilot script generation | Claude Opus 4.7 via OR | $0.40/script |
| Mood board visuals | kie.ai z-image | $0.02/board (5 images) |
| Casting brief generator | DeepSeek v3.2 | $0.02/brief |
| Budget framework calculator | GPT-5-mini via OR | $0.04/budget |
| Pre-viz scene (video) | Higgsfield Sora 2 / Veo 3 | $0.40/scene |
| Format-similarity check (anti-plagiarism) | OpenAI embeddings + Pinecone | $0.001/check |

**Per Format Bible total AI cost:** ~$2.10. Sell at $4,900. Even at high human-polish overhead the margin stays > 80%.

## 5. Cloudflare architecture
- **Worker:** `anekaria-com` — SSR landing + auth gate for studio dashboard
- **Worker:** `anekaria-api` — format generation queue + checkout webhook
- **Pages:** `anekaria-dashboard` — Studio Partner login (React/Vite)
- **D1:** `anekaria_db` — formats, clients, bibles, scripts, pilots, licensing
- **R2:** `anekaria-bibles` — generated PDF bibles + script docs + mood-board images
- **R2:** `anekaria-previz` — video pre-viz scenes (large files)
- **KV:** `anekaria-cache` — edge cache for marketing/showreel
- **Durable Objects:** `BibleSession` — multi-step bible generation state (40-page bible takes ~12 min)
- **Queue:** `anekaria-gen-q` — fan-out across Claude + kie + Higgsfield
- **Cron:** weekly newsletter aggregate + monthly Studio Partner report

## 6. 30/90 day plan
**Day 1–2 (now):** scaffold + brand kit + landing + 8 sub-pages + deploy
**Day 3–10:** Stripe + 3-tier checkout + simplest Format Spark generator (1 prompt → 5-page treatment PDF)
**Day 11–21:** Format Bible generator (multi-step Claude pipeline + PDF assembly + R2)
**Day 22–30:** First paying customers (Format Spark target: 5 buyers from LinkedIn outreach)
**Day 31–60:** Studio Partner pilot — 2 SEA OTT platforms, fully managed quarterly relationships
**Day 61–90:** Pre-viz video via Higgsfield + casting-brief LinkedIn integration + first format LICENSED

## 7. Moat
- **Format library compounds.** Every bible generated improves our prompts + few-shot examples. By month 6 we're top-of-class on format quality.
- **Network of producers.** Studio Partners refer each other. By year 2 we are the development-partner-of-record for 3+ OTT platforms.
- **AI-detection rights bundle.** We pre-clear format similarity (vs existing IP) → defensible deliverable that human-only studios can't match.
- **Speed.** Bible in 12 days vs 12 months. Anyone copying us needs the same AI stack + curation taste + producer network. Curation taste is hardest to copy.

## 8. Mandatory standards
- 100% Cloudflare. Mobile-native dashboard + landing. PWA. JSON-LD complete.
- T2I mood boards via kie.ai z-image. Logo via gpt-image-2 (3 concepts).
- Heroes piped via xmoj.com before R2/assets.
- WCAG AA on every text color (we'll do dark-mode landing — neon brights need verified contrast).
- a11y: bibles delivered as accessible PDFs (tagged, OCR'd, alt text).

## 9. Competitor synthesis
- **Endemol Shine / Banijay / Fremantle:** legacy format factories, slow + expensive. Win: 10x faster, 10x cheaper, AI-native.
- **Channel Factory / The Story Factory:** Western boutique studios, mostly UK/US focus. Win: SEA niche + Indonesian language + lower-priced tiers.
- **Adobe Firefly / Runway:** generic creative tools, not vertical for entertainment formats. Win: format-specialized prompts + producer network.
- **Solo creators on TikTok:** great at viral moments, can't deliver bible-grade IP. Win: structured deliverables platforms can commission.

## 10. Missing parts audit
- [x] Domain + CF zone
- [ ] OpenRouter API key with Claude Opus 4.7 access (Phase 2 — within $100/mo budget)
- [ ] Higgsfield Sora 2 / Veo 3 access for pre-viz (Phase 3)
- [ ] Stripe Indonesia merchant + multi-currency (Phase 2)
- [ ] PDF generation pipeline (pdfme / puppeteer; render bibles to branded PDF) — Phase 2
- [ ] LinkedIn Sales Navigator subscription for B2B outreach (Phase 2)
- [ ] 5 cornerstone case studies (fictional first, real after Format Spark customers ship) — Phase 3

---

## ⚙ HARD CONSTRAINTS (enforced for all sites)

This domain MUST operate within these constraints — no exceptions:

- **100% Cloudflare serverless** — Workers + D1 + R2 + KV + Workers AI + Vectorize. NEVER PM2, NEVER VPS, NEVER Docker in production path.
- **100% AI-automated** — every customer interaction, every moderation decision, every transaction reconcile = AI. No manual queue, no live human chat support, no physical fulfillment.
- **1-operator solo** — one person can run the entire operation from a phone. No team meetings, no shared inbox, no shift rotation.
- **WhatsApp AI bot** for all support (24/7, instant response, no SLA promises that need humans).
- **Mayar QRIS** for all Indonesian payments (subscription auto-renew, no manual invoicing).
- **Indonesian UI primary** — bahasa-first, English fallback only where unavoidable.
- **Privacy** — opt-in only, delete-on-request honored within 24h (cron-driven).
- **No physical goods, no inventory** — digital products + affiliate referrals only.

**If the plan above describes any flow that violates these constraints, treat the plan as ASPIRATIONAL only and rework before building. The constraint trifecta wins.**

PREVIEW

anekaria.com — Product Plan

AI Variety Show & Entertainment Format Studio. A creative studio that uses AI to generate, prototype, and produce entertainment formats — game shows, sketch series, talk show concepts, reality formats — for OTT platforms, broadcasters, and creator networks. Name = aneka (ID: variety) + studio energy.

1. The wedge — why anekaria wins

No category leader for "AI-native format development." Studios like Endemol Shine, Banijay, and FremantleMedia license formats but development is slow (12–18 months, $200K+ per format bible). Local production houses in SEA can't afford that pipeline. We compress it to weeks for one-tenth the cost.
AI is finally good enough for format ideation. GPT-5 + Claude 4.6 can produce defensible format treatments, pilot scripts, and visual mood boards in hours that previously took a writer's room weeks. Plus video pre-viz via Sora 2 / Veo 3.
Indonesian + SEA-first wedge. OTT platforms (Vidio, Viu, Vision+, Disney+ Hotstar, IDN Media) commission 200+ locally-relevant formats per year. None have an AI-native development partner. Win this niche → expand to MENA + LATAM.
Production-ready output, not just decks. Every format ships with bible + 3-episode pilot scripts + visual treatment + budget + casting brief — assembled by AI, polished by human producers. The "studio" not the "tool" positioning.

2. Money stack (3 tiers, USD priced)

Tier	Price	Output	AI cost	Margin
Format Spark	$499 one-time	1 format treatment (5pg) + mood board + 1 pilot scene	~$30	94%
Format Bible	$4,900 one-time	Full format bible (40pg) + 3 pilot scripts + visual treatment + casting brief + budget framework	~$300	94%
Studio Partner	$24,000/quarter	6 format bibles per quarter + monthly review + production support + format licensing rep	~$2,400 quarterly	90%

LTV proxy: Studio Partner repeat rate = 75% (production houses develop continuously). Format Bible upsells to Studio Partner = 38%.

Year-1 target: $400K (8 Studio Partners + 20 Format Bibles + 30 Format Sparks).

3. Acquisition (triple-stack)

LinkedIn long-form B2B — weekly essay aimed at heads of development at SEA OTT platforms. Hook: "Why your network commissions the same 4 formats every year." Build authority → inbound qualified leads.
Showreel via TikTok/Reels — 60-sec format pitches done in AI animation, shown publicly. Free creative samples that demonstrate quality. 1 reel per week.
Direct outreach to development heads at Vidio, Viu, Vision+, IDN, BRIO, GoPlay — handwritten-style intro email + a custom 3-format teaser already built for their slate gap.

4. AI automation map

Loop	Model	Cost guard
Format ideation (concept generator)	Claude Sonnet 4.6 via OR	$0.08/format
Format bible drafting (40-page)	Claude Opus 4.7 via OR	$1.20/bible
Pilot script generation	Claude Opus 4.7 via OR	$0.40/script
Mood board visuals	kie.ai z-image	$0.02/board (5 images)
Casting brief generator	DeepSeek v3.2	$0.02/brief
Budget framework calculator	GPT-5-mini via OR	$0.04/budget
Pre-viz scene (video)	Higgsfield Sora 2 / Veo 3	$0.40/scene
Format-similarity check (anti-plagiarism)	OpenAI embeddings + Pinecone	$0.001/check

Per Format Bible total AI cost: ~$2.10. Sell at $4,900. Even at high human-polish overhead the margin stays > 80%.

5. Cloudflare architecture

Worker: anekaria-com — SSR landing + auth gate for studio dashboard
Worker: anekaria-api — format generation queue + checkout webhook
Pages: anekaria-dashboard — Studio Partner login (React/Vite)
D1: anekaria_db — formats, clients, bibles, scripts, pilots, licensing
R2: anekaria-bibles — generated PDF bibles + script docs + mood-board images
R2: anekaria-previz — video pre-viz scenes (large files)
KV: anekaria-cache — edge cache for marketing/showreel
Durable Objects: BibleSession — multi-step bible generation state (40-page bible takes ~12 min)
Queue: anekaria-gen-q — fan-out across Claude + kie + Higgsfield
Cron: weekly newsletter aggregate + monthly Studio Partner report

6. 30/90 day plan

Day 1–2 (now): scaffold + brand kit + landing + 8 sub-pages + deploy Day 3–10: Stripe + 3-tier checkout + simplest Format Spark generator (1 prompt → 5-page treatment PDF) Day 11–21: Format Bible generator (multi-step Claude pipeline + PDF assembly + R2) Day 22–30: First paying customers (Format Spark target: 5 buyers from LinkedIn outreach) Day 31–60: Studio Partner pilot — 2 SEA OTT platforms, fully managed quarterly relationships Day 61–90: Pre-viz video via Higgsfield + casting-brief LinkedIn integration + first format LICENSED

7. Moat

Format library compounds. Every bible generated improves our prompts + few-shot examples. By month 6 we're top-of-class on format quality.
Network of producers. Studio Partners refer each other. By year 2 we are the development-partner-of-record for 3+ OTT platforms.
AI-detection rights bundle. We pre-clear format similarity (vs existing IP) → defensible deliverable that human-only studios can't match.
Speed. Bible in 12 days vs 12 months. Anyone copying us needs the same AI stack + curation taste + producer network. Curation taste is hardest to copy.

8. Mandatory standards

100% Cloudflare. Mobile-native dashboard + landing. PWA. JSON-LD complete.
T2I mood boards via kie.ai z-image. Logo via gpt-image-2 (3 concepts).
Heroes piped via xmoj.com before R2/assets.
WCAG AA on every text color (we'll do dark-mode landing — neon brights need verified contrast).
a11y: bibles delivered as accessible PDFs (tagged, OCR'd, alt text).

9. Competitor synthesis

Endemol Shine / Banijay / Fremantle: legacy format factories, slow + expensive. Win: 10x faster, 10x cheaper, AI-native.
Channel Factory / The Story Factory: Western boutique studios, mostly UK/US focus. Win: SEA niche + Indonesian language + lower-priced tiers.
Adobe Firefly / Runway: generic creative tools, not vertical for entertainment formats. Win: format-specialized prompts + producer network.
Solo creators on TikTok: great at viral moments, can't deliver bible-grade IP. Win: structured deliverables platforms can commission.

10. Missing parts audit

Domain + CF zone
OpenRouter API key with Claude Opus 4.7 access (Phase 2 — within $100/mo budget)
Higgsfield Sora 2 / Veo 3 access for pre-viz (Phase 3)
Stripe Indonesia merchant + multi-currency (Phase 2)
PDF generation pipeline (pdfme / puppeteer; render bibles to branded PDF) — Phase 2
LinkedIn Sales Navigator subscription for B2B outreach (Phase 2)
5 cornerstone case studies (fictional first, real after Format Spark customers ship) — Phase 3

⚙ HARD CONSTRAINTS (enforced for all sites)

This domain MUST operate within these constraints — no exceptions:

100% Cloudflare serverless — Workers + D1 + R2 + KV + Workers AI + Vectorize. NEVER PM2, NEVER VPS, NEVER Docker in production path.
100% AI-automated — every customer interaction, every moderation decision, every transaction reconcile = AI. No manual queue, no live human chat support, no physical fulfillment.
1-operator solo — one person can run the entire operation from a phone. No team meetings, no shared inbox, no shift rotation.
WhatsApp AI bot for all support (24/7, instant response, no SLA promises that need humans).
Mayar QRIS for all Indonesian payments (subscription auto-renew, no manual invoicing).
Indonesian UI primary — bahasa-first, English fallback only where unavoidable.
Privacy — opt-in only, delete-on-request honored within 24h (cron-driven).
No physical goods, no inventory — digital products + affiliate referrals only.

If the plan above describes any flow that violates these constraints, treat the plan as ASPIRATIONAL only and rework before building. The constraint trifecta wins.

AI ASSISTANT

Ask AI to research, improve, or generate content.

Try: "Research competitors for this niche"

Actions