changelog-video

Converts weekly changelog markdown into a branded square video with voiceover, captions, and animated UI mockups.

13|17|Updated Jan 30, 2025
One-click install
npx skills add https://github.com/CodeWithSally/CodeWithSally-Apex-English --skill changelog-video-codewithsally
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: changelog-video
Source: https://github.com/CodeWithSally/CodeWithSally-Apex-English/tree/main/Sessions/ClaudeCode/Session007/.agents/skills/changelog-video
Command: npx skills add https://github.com/CodeWithSally/CodeWithSally-Apex-English --skill changelog-video-codewithsally

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Turning a weekly changelog into a polished branded video normally requires manual scripting, voiceover recording, motion design, and caption timing. This Skill automates that entire pipeline, producing a lint-clean HyperFrames project from a single changelog markdown file. ## Core Features & Use Cases - End-to-end video pipeline: Parses a changelog digest, cuts it editorially to a 45-60s script, generates Annie voiceover via HeyGen TTS, and builds a 1080x1080 branded composition with animated background, fonts, and caption rail. - Two-layer script with phonetic lexicon: Technical terms (JSON, CLI, GSAP) carry separate spoken and display forms so the TTS pronounces them correctly while captions show standard spelling, with automated timestamp alignment via align-captions.mjs. - Visualization-first scenes: A routing registry maps each changelog theme to an animated UI mock, terminal demo, or checklist instead of bullet-point slides. - Use Case: Given the weekly HyperFrames digest markdown, produce a ready-to-render square video project with voiceover, word-timed captions, and brand-consistent motion, verified through lint and seam gates. ## Quick Start Use the changelog-video skill to turn this weekly changelog markdown into the branded weekly video project.

Frequently Asked Questions about changelog-video

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

FAQPage Schema
How do I turn a changelog into a video automatically?

Provide a changelog markdown with themes and items, and the skill parses it, writes a two-layer script, generates HeyGen voiceover, aligns caption timestamps, and builds a branded HyperFrames composition. Rendering only happens when explicitly requested.

How do I make TTS pronounce technical terms like JSON or CLI correctly?

The skill uses a lexicon mapping display spellings to phonetic spoken forms, such as JSON to jay-sawn and CLI to C L I. Captions keep standard spelling while the voiceover reads the phonetic form, and unknown terms must be added to the lexicon rather than guessed.

What happens if the TTS returns no word timestamps?

Empty word-timestamp output is a hard gate; the build cannot proceed without it. The fallback is forced alignment of the produced audio with local whisper, then running align-captions.mjs against that output so captions still ship.

Can I use my own fonts or background instead of the bundled assets?

No, the skill treats its bundled fonts, background pattern, and BGM as the brand definition. Bootstrapping from the master-skeleton scaffold and copying the skill's assets verbatim is mandatory; hand-written font-face rules or custom backgrounds are explicit anti-patterns.

Why does my changelog video render without captions?

Missing captions mean the caption rail's LINES array was left empty in the scaffold. Populate it from captions.json produced by align-captions.mjs, then verify sampled frames show visible caption text before shipping.