Learn how a real AI video content pipeline is built — not a toy demo.

This is a reference build, not a one-click template. If you want to understand how experienced builders architect a full idea-to-published-video AI pipeline in n8n — including human approval loops, multi-provider fallback, and resumable state tracking — this is for you. If you want something that runs with zero setup on day one, this is not it (see Requirements below).

What this pipeline does

Generates original AI animal "POV" video concepts → routes them through a Telegram approval step → writes a 7-scene script → generates character-consistent stills (Replicate FLUX 1.1 Pro Ultra) → animates them into clips (Runway ML / Replicate Kling) → generates matching sound effects (ElevenLabs) → renders the final cut (Creatomate) → writes an SEO description and publishes to YouTube automatically. Every step's status is tracked in Airtable, so a failed run can resume instead of restarting from scratch.

What's included

• - Full sanitized n8n workflow JSON (117 nodes, 8 modules)

• - 9 documentation files — one per module, explaining the "why," not just the "what"

• - Airtable base schema so you can recreate the control-plane database

• - Cost breakdown for every external service used

• - Sample test payload to try Module 1 in isolation before touching the full pipeline

How to use it

1. 1. Read docs/00-architecture-overview.md first — do not import and run blind

2. 2. Recreate the Airtable base using the provided schema

3. 3. Connect your own credentials for each service you want to test (start with Module 1 only)

4. 4. Work module-by-module — each doc has an isolated test path

Requirements — read before buying

This workflow calls 8 paid third-party services: OpenRouter, Replicate, Runway ML, ElevenLabs, Creatomate, Google Cloud (YouTube + Drive), Airtable, and Telegram. You need your own accounts and API keys for all of them. Estimated per-video cost is detailed in docs/cost-breakdown.md. Requires n8n 1.70+ and the @n8n/n8n-nodes-langchain community package if self-hosted.

Refund policy

You may request a full self-service refund within 14 days of purchase if you have not downloaded the files. After downloading, contact support via Report a problem. See our Help Center for details.

FAQ

1. 1. Is this plug-and-play?

No. This is a reference/educational build for buyers who want to study a production-grade architecture. You'll need to connect 8 third-party services yourself. See the cost breakdown and requirements before buying.

2. 2. Why should I pay when there's a similar free template on n8n.io?

The free official template covers the general shape. This product adds a documented approval-loop UX, a resumable Airtable state-machine pattern, per-module isolation testing, and a sanitized, credential-safe export you can actually study node-by-node. If you just want "a" video pipeline with no explanation, the free option may be enough for you — we're transparent about that.

3. 3. Do I need to know how to code?

No coding required, but comfort with n8n (expressions, credentials, HTTP Request nodes) is assumed. This is not a beginner-first product.

4. 4. What does it cost to run per video?

See docs/cost-breakdown.md in the download — costs vary by provider usage but expect several dollars per finished video given the number of paid AI generation calls involved.

5. 5. What if I only want part of the pipeline?

Each module is documented and testable independently — most buyers start with Module 1 (idea + approval loop) or Module 3 (image generation) and adapt from there.

6. 6. Refund policy?

Full self-service refund within 14 days if files have not been downloaded. After download, contact support via Report a problem.