/brag
Turns the project you just built into a short, shareable launch video with one command, reading your code directly instead of asking for a live URL.
Overview
/brag is a Claude Code skill that turns a project you just built into a short, polished launch video, complete with music, motion, and share copy, powered by Hyperframes. It reads your project code directly, so there is no live URL or screenshot needed, and it also works with Codex CLI, opencode, Google Antigravity, and any agent that supports custom instructions.
How it works
- It reads the current project code to understand what the app actually does.
- It plans a short, specific angle for the video and writes a beat by beat storyboard, including scenes, timing, transitions, and sound cues.
- It hands a focused composition brief to Hyperframes, which builds, times, and renders the video.
- It validates the render, picks the best poster frame, and writes a
brag-output/folder with the plan, the composition, the renderedbrag.mp4, and share copy ready to post.
Examples
let's /brag
/brag --tone "fake Series A launch from 2016"
/brag --voice
Voiceover is off by default. Passing --voice turns on narration through Kokoro via Hyperframes.
Installation
You need an agent that supports Agent Skills (Claude Code, opencode, Codex CLI, or similar), Node.js 22 or newer, FFmpeg on your PATH, and the Hyperframes CLI.
- In Claude Code, install the plugin:
/plugin marketplace add latent-spaces/bragthen/plugin install brag@brag. - For any other supported agent, run
npx skills add https://github.com/latent-spaces/brag --skill braginstead. - Check that Hyperframes is ready with
npx hyperframes doctor. - From inside any project, ask your agent
let's /brag. - Verify it worked by checking the new
brag-output/folder forbrag-plan.md, the composition files, and the renderedbrag.mp4.
Related assets
Frontend Design
Generates distinctive, production-grade UI that avoids generic "AI slop," with bold typography, cohesive color, and precise motion, implemented as real code.
Graphify
Turns a codebase, plus its docs, PDFs, images, and video, into a queryable knowledge graph that a coding agent can traverse instead of grepping raw files, and never guesses silently: every edge is tagged EXTRACTED, INFERRED, or AMBIGUOUS so you know what was found versus inferred.
Caveman
Makes your coding agent answer in a stripped down, high signal style that keeps every technical fact and cuts filler, saving output tokens on chat style questions.