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 Sora 2, Veo 3.1, Kling 3.0, and Nano Banana.

Get started (5 minutes)

1. Clone this repo

git clone <repo-url>
cd arcads-agent-skills

2. Run setup

./scripts/setup.sh

This will:

• 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. The agent loads the Arcads skill automatically.

Cursor: Open the folder. The skill is at .cursor/skills/arcads-external-api/.

4. Start creating

The agent handles API calls, polling, prompt engineering, and file organization. Here are the main workflows:

Create an AI influencer (character sheet)

"Create a new AI influencer — a 22-year-old college student with freckles"

The agent generates a full-body hero image for your approval, then creates 9 additional angles (3/4 views, profile, closeup, etc.) using the hero as a reference. All 10 images are saved to references/influencers/ for future use.

Generate UGC product selfie stills

"Generate a UGC selfie of Sofia holding the Arcads Cola can in her bedroom"

Combines your character + 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.

Animate a still into video

"Turn that image into a video — have her talk about the product"

Uses Veo 3.1 with startFrame to animate your approved UGC still. The video starts from that exact image with natural human motion (eye contact breaks, head tilts, body shifts) and dialogue.

Quick UGC video (no starting frame)

"Generate a UGC video ad for this product" + drop a product photo

Uses Sora 2 with your product photo as a style reference to generate a video directly — no starting frame needed. Faster but less control over the person's appearance.

Other things to try

• "Recreate this influencer's look from a reference photo"
• "Make a Nano Banana product hero image"
• "Generate 5 different ad variations for this product"

What's in the box

Path	What it does
skills/arcads-external-api/	The skill: API reference, prompting guide, per-model prompt library
MASTER_CONTEXT.template.md	Template for your workspace context (credit costs, brand voice, learnings)
MASTER_CONTEXT.md	Your personalized copy (created by setup, not committed to git)
.env	Your API key (created by setup, never committed)
scripts/setup.sh	One-time setup
scripts/sync-skill.sh	Copies skill edits to .claude/ and .cursor/ directories
scripts/check-arcads-env.sh	Tests API connectivity
references/	Drop reference images here (influencers, products, aesthetics) — gitignored

Your API key

Your key authenticates with the Arcads API. During setup you paste it once and the agent uses it from .env automatically. You never need to paste it into chat.

Find your key: Arcads Dashboard > Settings > API

Project memory

MASTER_CONTEXT.md is your workspace's living memory. The agent reads it at the start of every session and writes learnings back. It stores:

• Default product — auto-populated on first use so you're never asked "which product?" again
• Credit costs — you fill in once (or the agent asks), then every session has them
• Brand voice — optional tone, audience, and word preferences
• API learnings — universal Arcads quirks that help the agent work better
• Changelog — dated notes from each session

Supported models

Model	Type	Best for	Credits
Veo 3.1	Video	Animating a starting frame into ~8s video with dialogue. Best for UGC stills → video.	1
Sora 2	Video	Longer videos (up to 20s) from text prompts. Product photo as style ref (no starting frame).	varies
Kling 3.0	Video	B-roll and scene generation (via scene/b-roll endpoints)	varies
Nano Banana 2	Image	UGC stills, character sheets, product shots, influencer recreation	0.03

Reference images

Drop images into the references/ folder and the agent will use them automatically:

• references/influencers/ — Photos of people to recreate as AI-generated content
• references/products/ — Product photos for showcase videos and hero images
• references/aesthetics/ — Style references organized by vibe (ugc-selfie/, cinematic/, etc.)

Images stay local — the folder contents are gitignored.

Editing the skill

The canonical skill source lives in skills/arcads-external-api/. After editing any file there, run:

./scripts/sync-skill.sh

This copies your changes to .claude/skills/ and .cursor/skills/ (which are gitignored — they're generated copies).

Security

• .env is gitignored — never committed
• MASTER_CONTEXT.md is gitignored — contains your product IDs and workspace data
• Never paste API keys in GitHub issues or public chats

Vendor prompting guides

Model	Guide
Sora 2	OpenAI — Sora 2 prompting guide
Veo 3.1	Google Cloud — Veo 3.1
Kling 3.0	Kling — user guide
Nano Banana	Google Cloud — Nano Banana

API docs

Arcads Swagger UI

Other AI assistants (Manus, Copilot, etc.)

Point your assistant at AGENTS.md and MASTER_CONTEXT.md + the skill path. See AGENTS.md for details.