pr-to-video

Converts GitHub pull requests into narrated code-change explainer videos via HyperFrames.

Updated Sep 6, 2026
One-click install
npx skills add https://github.com/AbuBakar125-co/Resturant_web --skill pr-to-video-abubakar125-co
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-to-video
Source: https://github.com/AbuBakar125-co/Resturant_web/tree/main/agent/skills/pr-to-video
Command: npx skills add https://github.com/AbuBakar125-co/Resturant_web --skill pr-to-video-abubakar125-co

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Turning a merged or proposed pull request into a clear video explanation normally requires manual screen recording, scripting, and editing; this Skill automates that pipeline by reading the PR's diff, commits, and contributors and producing a finished MP4 explainer. ## Core Features & Use Cases - Deterministic PR ingestion: Uses the GitHub CLI (gh) to fetch PR metadata, the full diff (paginated for large PRs), and contributor avatars, with no website capture step. - End-to-end video pipeline: Orchestrates setup, storyboard and script drafting, TTS narration with word timings, BGM/SFX retrieval, frame-by-frame HTML composition building, caption generation, and final MP4 rendering through the HyperFrames CLI. - Code-editorial design system: Applies a fixed code-editorial preset with a navy code surface, rendering real diff hunks through code-* registry blocks (diff, morph, typing) and closing with a contributor credits frame. - Use Case: After merging a feature PR like acme/sdk#1842, ask the agent to turn that PR into a 60-second changelog video; it fetches the diff, drafts a storyboard for approval, generates narration and music, and renders renders/video.mp4. ## Quick Start Turn the pull request at https://github.com/owner/repo/pull/123 into a short code-change explainer video with narration and background music.

Frequently Asked Questions about pr-to-video

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a GitHub pull request into a video?

Provide a PR URL, an owner/repo#N reference, or run it in a checked-out repo with 'this PR'. The skill fetches the diff and metadata via the gh CLI, drafts a storyboard and script for approval, generates narration and music, builds HTML frames, and renders an MP4.

What inputs does the PR to video workflow accept?

It accepts a full GitHub PR URL, a shorthand owner/repo#N reference, or the phrase 'this PR' inside a checked-out repository. It does not capture websites; the only real assets are contributor avatars downloaded from GitHub.

Does it work with large pull requests with many files?

Yes. The fetch-pr script completes the files list through paginated gh api calls, so a large PR does not truncate at roughly 100 files. For merged PRs it also resolves a best-effort shipped version for the end card.

Can I make a silent video without narration or music?

Yes. Set music: none in the storyboard's top YAML block and omit SCRIPT.md; the audio step recognizes this as the fully-silent marker and generates nothing. Using music: none with a script keeps narration but disables only the background music.

Why did the PR fetch step fail with an authentication error?

fetch-pr.mjs exits with an error when gh is not authenticated, the PR is not found, or the repository is private without access. Run gh auth login or adjust repository permissions, then rerun; the skill never fabricates PR contents.

When should I use a different video skill instead?

Use a product-launch video skill for product promos and a faceless-explainer skill for topic explainers without a PR. This skill is specifically for code changes; unclear intents route through the hyperframes intent layer.