arcads-claude-code

Arcads AI Video — Agent Skill Pack

Create AI marketing videos and images using your Arcads account, powered by AI agents in Claude Code or Cursor. Supports the full Arcads creative stack — Seedance 2.0 (flagship video), Sora 2, Veo 3.1, Kling 3.0, Grok Video, Nano Banana 2 / Pro / Edit, ChatGPT Image 2, OmniHuman, and Audio-driven — plus a 37-template static Meta image-ad library and a pipeline for Pixar-style and claymation animated ads.

🎥 Watch the walkthrough

How To Build Entire AI Ad Campaigns With Claude Code (free repo) — full build tutorial by Mr. Paid Social.

Level up your media buying with AI

This repo handles the creative side. If you want the full system — scaling, ROAS, automation, and the AI workflows behind 8-figure ad accounts — come build with me inside The AI Ad Alchemists.

It's a private Skool community of 460+ media buyers managing 8-figure ad accounts, run by me (Caleb — aka "Mr. Paid Social") drawing on 12 years and $150M+ in ad spend. Inside you get:

• Meta Masterclass — the exact systems I use to scale Meta ads (valued at $1.2k)
• Custom GPTs for ad copywriting and compliance
• Ad swipe files + breakdowns of what's working right now
• Airtable & Google Sheets scaling systems — the operational backbone behind 8-fig accounts
• AI tool walkthroughs — including the Ad Agent (built in Claude Code) and the GenAI system in Airtable I'm shipping next
• Monthly group calls + guest speakers
• Direct access to me and the network

If you're using this repo to crank out creative, the community is where you learn to turn that creative into ROAS at scale.

→ Join The AI Ad Alchemists — $97/month

Facebook Ad Library → Airtable swipe file

Build a searchable library of competitor image ads with the included Meta Ad Library → Airtable workflow. Supply up to 10 advertiser or brand entries; the default collection limit is 100 active image ads per entry, subject to what Meta's API returns.

The workflow collects ad metadata through Meta's Ad Library API, then uses authorized automated headless Chrome to retrieve images from the API's snapshot links. An agent or person reviews each downloaded creative before it is imported into Airtable with:

• Image and thumbnail attachments, plus the public Ad Library link.
• Brand and advertiser details, ad copy, dates, platforms and verified CTA links.
• Relevant creative tags and format choices matched to your table's schema.
• An individual recreation prompt describing the image's layout, typography, colors, subject and text hierarchy for your chosen image generator, such as ChatGPT Image 2.5.

Ad-ID checks skip previously imported ads, and the importer verifies that Airtable has received the attachments. The advertised brand stays separate from the Facebook Page, so a shared advertiser such as Skool can be filtered into the right communities.

Try this with your agent:

"Follow scripts/meta-ad-library-import/README.md. Import up to 100 active static image ads for each of these Facebook Page IDs: [IDs], into [Airtable URL]. Use the credentials in my .env. I authorize automated headless extraction of the API-provided snapshots. Inspect every image, fill all applicable fields, select relevant existing tags and formats, and write an individual ChatGPT Image 2.5 recreation prompt. Check for duplicate ad IDs and verify the uploaded attachments."

You need a Meta token with Ad Library access, an Airtable PAT or connected Airtable MCP, and Chrome with the extraction dependencies. API coverage varies by delivery country and access; the tested setup uses explicit UK/EU countries. This workflow writes prompts and does not generate replacement ads.

Full setup and commands · Agent handoff guide · Advertiser example

Prerequisites

The agent and the basic Arcads workflows (image generation, video generation, polling) work with just Python 3.10+ and the API key from setup. Some multi-step pipelines need a few extra CLI tools:

Tool	Required for	Install (macOS)
Python 3.10+	Everything (the image-ad generators are stdlib-only — no pip install)	preinstalled or brew install python@3.12
ffmpeg	Pixar-style ad, claymation ad, caption-video (stitching + chroma-key overlay)	brew install ffmpeg
jq	Several bash scripts (pixar-style-ad, etc.)	brew install jq
Node.js + npx hyperframes	Caption burn-in workflow	brew install node (the skill runs npx on demand)
whisper Python package	Caption transcription	pip install openai-whisper (or pip3)
meta-ad-builder deps	Publishing to Meta Marketing API	pip install -r shared/skills/meta-ad-builder/scripts/requirements.txt
Chrome + Selenium/Pillow	Meta Ad Library image extraction	Chrome, then install workflow dependencies in a virtual environment

The image-ad generator scripts (chatgpt-image-ad, nano-banana-image-ad, image-ad-clone) are intentionally stdlib-only — no extra installs needed. The deps above are only required when you invoke the matching multi-step workflow.

Linux users: apt install ffmpeg jq nodejs python3. Windows users: WSL2 recommended; the shell scripts assume bash.

Get started (5 minutes)

1. Clone this repo

git clone https://github.com/krusemediallc/arcads-claude-code.git
cd arcads-claude-code

2. Run setup

./scripts/setup.sh

This will:

• Sign up if you need an Arcads account: arcads.ai/?via=claude-code
• Ask for your Arcads API key (find it at app.arcads.ai/settings/api)
• Save it securely in .env (never committed to git)
• Verify your connection to Arcads
• Create your personal MASTER_CONTEXT.md workspace file

3. Open in your AI editor

Claude Code: Open the folder. A SessionStart hook runs and prints an orientation banner showing which skills are installed, whether your .env and MASTER_CONTEXT.md are set up, and where the docs live.

Cursor: Open the folder. Same skills are exposed at .cursor/skills/.

4. Start creating

The agent handles API calls, polling, prompt engineering, file organization, and cost confirmation. Workflows are grouped by what you want to make.

---

🎬 Seedance 2.0 videos (the flagship — start here)

Seedance 2.0 is the most flexible model in the stack — 4–15s clips, native audio, image-to-video or video-to-video, reference images, multiple shot styles. Five prompt formulas ship with the skill:

UGC selfie-style product review

"Make a 12-second Seedance UGC video — woman in a kitchen, holding the product, says she stopped buying [competitor]"

Uses the 9-layer UGC formula tuned for Seedance 2.0 (iPhone-shot aesthetic, natural eye-contact breaks, casual delivery). See skills/arcads-external-api/prompting/prompt-library/seedance-2-ugc.md.

Premium product reveal (no person)

"Premium reveal of [product] — dark void, text narrative, hero rotation"

Dark-void aesthetic, text overlays narrating the product's positioning, no person on screen. See skills/arcads-external-api/prompting/prompt-library/seedance-2-premium-reveal.md.

Product hero with elemental effects

"Seedance product hero — water splash, mist, slow rotation"

Splash, mist, light rays, slow rotation. See skills/arcads-external-api/prompting/prompt-library/seedance-2-product-hero.md.

Studio lookbook with voiceover

"Studio lookbook of [product] — multi-look, polished, with voiceover script"

Polished editorial / lookbook style, multi-shot, with embedded dialogue. See skills/arcads-external-api/prompting/prompt-library/seedance-2-studio-lookbook.md.

Feature walkthrough demo

"Seedance feature walkthrough — fast-paced, show off [features]"

Fast-paced product-demo cuts. See skills/arcads-external-api/prompting/prompt-library/seedance-2-feature-walkthrough.md.

---

🎬 Other video models

Sora 2 (text-to-video, up to 20s)

"Generate a 16-second Sora video of [scene]" — optionally drop a product photo as a style reference.

Sora 2 handles longer durations than Veo. The agent auto-selects duration from your script's word count (~2.5 words/sec). Sora 2 remix is also supported via POST /v1/sora2/remix/video for remixing an existing asset.

Veo 3.1 (starting-frame animation)

"Animate this Nano Banana still into an 8-second Veo with dialogue"

Veo 3.1 with startFrame is the standard path for UGC stills → video. The video starts from your exact image with natural human motion and embedded dialogue. The agent confirms the dialogue separately before generating (the MANDATORY dialogue gate).

Kling 3.0 (b-roll / scene)

"Make a 5-second b-roll clip of [scene]" or "Generate a scene of [environment]"

Kling 3.0 is the b-roll / scene workhorse. Hits the dedicated POST /v1/b-roll and POST /v1/scene endpoints.

Grok Video

"Generate a Grok video of [scene]"

Hits POST /v2/videos/generate with model: "grok-video".

OmniHuman / Audio-driven

"OmniHuman avatar of [person] delivering [script]" or "Audio-driven video lip-synced to [audio file]"

POST /v1/omnihuman for talking-avatar workflows; POST /v1/audio-driven for lip-sync against an audio file.

---

🖼️ Image generation (Nano Banana + ChatGPT Image 2)

Create a new AI influencer (10-image character sheet)

"Create a new AI influencer — 22-year-old college student with freckles, golden-hour kitchen lighting"

Two-pass workflow: (1) generate a hero front portrait via Nano Banana, get your approval, (2) generate 9 additional angles (3/4 views, profile, close-up, expressions) with the hero as the reference. All 10 saved to references/influencers/ for future reuse.

UGC product selfie still

"Generate a UGC selfie of Sofia holding [product] in her bedroom"

Combines your character hero + product photo + style references from references/aesthetics/ugc-selfie/ into an authentic-looking iPhone selfie frame grab. Includes skin realism and camera imperfections to fight AI's polished default.

Product showcase still (Nano Banana → Veo / Seedance video)

"AI person holding [product] talking about [feature]"

Two-step: Nano Banana still of person + product → user approves → start-frame → video via Veo 3.1 or Seedance 2.0.

Recreate an influencer from a reference photo

"Recreate this influencer's look from this reference photo"

Two-step: Nano Banana still from refImageAsBase64 → user approves → Veo 3.1 with startFrame.

Nano Banana model choice

Default is nano-banana-2. Use model: "nano-banana" for Nano Banana Pro (Gemini 3 Pro Image — higher fidelity, locks character identity tighter across reference batches). nano-banana-edit for inpainting.

---

📸 Static Meta image ad creative (37-template library)

"Make me an Apple Notes-style ad for my product" / "Generate a Forbes editorial ad" / "Clone this comparison-table ad as a template"