Analysis updated 2026-05-18
Have an AI agent produce a brand-matched 15 second promotional video ad from a JSON brand kit.
Reuse the same video ad template across multiple brands by swapping in different brand data.
Follow a structured research-interview-brief-build-review workflow for AI-assisted ad production.
Generate ad images, voiceover, and music automatically through an AI asset generation service.
| loopsera/agentic-ad-studio | 0xradioac7iv/tempfs | 7vignesh/pgpulse | |
|---|---|---|---|
| Stars | 0 | 0 | 0 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 3/5 | 4/5 |
| Audience | vibe coder | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires a kie.ai API key for asset generation and Node.js for the Remotion video project.
Agentic Ad Studio is a template and workflow for having an AI agent produce short, brand-matched video ads. Instead of a designer manually building each ad, the repo defines a step by step process, called an SOP, that an agent follows: research the brand, interview the user about goals, write a creative brief, build the ad, get it reviewed, then render the final video, with checkpoints along the way where a human approves before moving on. Brand identities are stored as plain JSON files describing colors, fonts, and copy. The actual video templates are built with Remotion, a tool that lets you create videos using React code, so the same template logic can be reused across different brands by swapping in a different brand file. Images, voiceover, and background music for each ad are generated using AI through a service called kie.ai. The repo already includes five example ads built this way, for Google Gemini, the project's own brand Loopsera, Proton VPN, Uber, and Airbnb, kept as demo and portfolio pieces rather than official ads for those companies. To use it, a person clones the repository, opens it in Claude Code, which reads the included project notes to understand the setup, and follows the documented workflow file to produce an ad for a new brand. Setup involves installing Node.js dependencies for the video project, downloading the needed fonts, which are not included directly for licensing reasons, and adding an API key for the asset generation service. Each brand gets its own render command that outputs a finished video file. The README notes that Remotion itself is free for individuals and small teams but requires a paid license for companies with four or more people, and that fonts used for well-known brands are fetched from public sources rather than bundled. No overall license for this repository itself is stated in the README.
A template where an AI agent follows a checkpointed workflow to research, design, and render short brand-matched video ads using Remotion.
Mainly TypeScript. The stack also includes TypeScript, React, Remotion.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.