Communitygithub.com

josueh04/product-video-skills

Coordinate the build of one or more signed videos of the current product with subagents (source recon, product truth, UI specs, voice, one builder per video), re-run QA itself, then deliver. A light coordinator that never builds itself. Run only when the user types /video-build.

product-video-skills 是什么？

product-video-skills is a Claude Code agent skill that coordinate the build of one or more signed videos of the current product with subagents (source recon, product truth, UI specs, voice, one builder per video), re-run QA itself, then deliver. A light coordinator that never builds itself. Run only when the user types /video-build.

兼容平台✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/josueh04/product-video-skills/tree/HEAD/skills/video-build
查看原文→浏览所有技能

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

OpenAIChatGPTClaudeClaudeGeminiGeminiGrokGrokPerplexityPerplexityDeepSeekDeepSeek

文档

/video-build

You are the coordinator. You check the gate, dispatch subagents stage by stage, verify each stage by its files, re-run QA yourself, deliver, and report. You do not write composition code, specs or narration, and you do not render.

Why the split: in the production these skills come from, while the main agent built videos itself its context filled up 8 times, once in the middle of a fix. When it only coordinated (consolidating, writing briefs, reviewing evidence, delivering), three fixes shipped in 70 minutes. Coordination is also where the quality gate lives: subagents reported "clean" renders that the coordinator's own QA then caught with a 0.3 s flash and a clipped modal.

0. Find the work

PVS_HOME="$(cd "$(cd "${CLAUDE_SKILL_DIR}" && pwd -P)/../.." && pwd)"
PY="$PVS_HOME/bin/pvs-py"; S="$PVS_HOME/skills/video-build/scripts"   # quote both when used
  • With <video>: videos/<video>/ in the current product.
  • Without it: the video folder you are in, or else every video of the product whose sheets are signed and whose current version has no delivery. Show that list and confirm it in one closed question before dispatching anything.

Run "$PY" "$S/stages.py" <video_dir> for each video and show the table. Stages that are already done are skipped, so /video-build can resume a build that stopped halfway.

1. The gate

"$PY" "$S/stages.py" <video_dir> --require signoff. If either signature in BRIEF.md is empty, stop and tell the user what is missing and how to sign (the reviewer writes their name, or tells you to run sign.py from /video-new). This is the check that keeps the costliest kind of rework (a feature the reviewer wanted explained and nobody wrote down) from starting a build.

--draft is the one exception, and only when the user typed it: a draft can be built, rendered and QA-checked so the reviewer sees something before signing, but deliver.py refuses it. Say so in the report.

Also read product.yaml, kit/RULES.md, BRIEF.md, COVERAGE.md and CLAIMS.md yourself: you need them to write the subagent prompts and to judge their results. Read nothing else in depth.

2. Dispatch

Use the Agent tool with background subagents. Prompts for every stage are in references/stage-prompts.md and the builder prompt is references/builder-prompt.md; read them before the first dispatch and fill every placeholder. After each subagent reports, run "$PY" "$S/stages.py" <video_dir> --require <stage> before starting anything that depends on it.

WaveStageSubagentsNeedsWrites
1recon1 per video, source-reconsignoffsources.lock, SOURCES.md
2truth1 per video, product-truthreconTRUTH.md
2specs1 per screen, ui-spec-from-code (or ui-reference-capture for screens without code)reconspecs/<screen>.md
3voice1 per video, script-and-voicetruthaudio/lines.tsv, clips/, timings.json
4compose + render1 builder per video, ui-demo-composer + seek-safe-motionspecs, voicevideo/, renders/<video>-v<N>.mp4
5qayourenderqa/REPORT.json
6deliveryouqadeliveries/

Waves run in parallel across videos and within a wave (truth and the specs of every screen go out together). Several videos share one machine: say in every prompt that no agent kills a process it did not start and that renders may be slower while others run.

Before wave 3 on version 1, if the voice provider is paid, show the user the narration plan in one message: the chapter table, the tagline, the voice, at most three closed decisions with defaults. Recording narration spends shared quota, and a tone or a name the reviewer dislikes is cheapest to change before it is recorded. If the user already said to go straight through, skip this and list the choices for veto in the final report instead.

While subagents run, do not poll them with messages. Check progress through files (modification times, $STAGES, the tail of a render log) and answer their questions when they send them. Approvals go up to the user, decisions come back down: if a subagent asks for something only the user can allow (a download, a new asset source), ask the user one closed question and relay the answer.

If a subagent stops responding, start a new one with the same prompt plus "previous agent stopped at ; continue from the files". State lives in the files, not in the agents.

3. Your own QA, whatever the builder reported

"$PVS_HOME/bin/pvs-py" "$PVS_HOME/skills/render-qa/scripts/qa.py" <video_dir> <mp4> \
  --strip <name>:<t0>:<t1> ...

Load the render-qa skill for how to read the result. Add a strip around every chapter title, every modal and the end screen to lockup handoff (times from the builder's chapter table). Then look yourself at qa/REPORT.md, every contact sheet and every strip. Check the narration against COVERAGE.md: each "yes" row must be explained in its chapter.

If anything fails, send the exact problem back to the builder (SendMessage while it is alive, the condensed fix format in references/stage-prompts.md), or, if it is gone, start a fixer with the prompt in $PVS_HOME/skills/video-review/references/fix-prompt.md (read the file; do not invoke that skill, it is user-invoked). Re-run this step on the new render. Never deliver a render you did not QA yourself.

4. Deliver

"$PVS_HOME/bin/pvs-py" "$PVS_HOME/skills/render-qa/scripts/deliver.py" <video_dir> <mp4> --archive-previous

It refuses a draft, a failed report, or a report whose sha256 does not match the file. Do not work around a refusal: fix the cause. Never overwrite a delivered file; deliver.py bumps the name instead and warns, and you say so.

5. Report

One table for the whole batch, then at most three closed decisions:

VideoVersionDurationChaptersCoverage explainedQADelivered as

Below it: what is real and what is assumed, anything unverified, the names, voices and tagline for veto, and the next step (/video-review with the reviewer's notes). Write in the user's language, short, tables first.

When to read the references

  • references/stage-prompts.md: before dispatching waves 1 to 3, and for any fix request.
  • references/builder-prompt.md: before dispatching wave 4.

Individual skills in this repo

This repo contains 14 individual skills — each has its own dedicated page.

josueh04/product-video-skills

Extract, once per product, everything every video of it reuses and write it to kit/ and product.yaml (design tokens, font subsets as woff2, icon subsets as SVG from the product's own icon packages, logos in light, dark and app-tile variants from the repo, a fictional cast proposed once for veto and then frozen, the canonical-names map, pronunciations, banned terms and the read-only tool list). Use it when a product is set up or its kit is missing or incomplete, when a video needs an icon, font or logo that is not in kit/ yet, when someone asks for demo names, fake customers, phone numbers or emails, when a brand word is mispronounced or an old product name shows up, and when checking that demo data is fictional.

josueh04/product-video-skills

Interview the user about a product, then create products/<slug>/ with its own git history, a filled product.yaml and linked skills, fetch its sources and build its kit. Run only when the user types /product-new.

josueh04/product-video-skills

Back every sentence of a video's narration (audio/lines.tsv) and every screen it shows with a citation into the pinned source code (role/path:line@sha) or a docs URL, mark what is visible in the UI versus backend-only, flag restricted or unreleased features, and cut or rewrite anything unbacked; writes the video's TRUTH.md and checks it with truth_check.py. Use it whenever a script or narration is drafted or edited, before voice is generated, before a build, when someone asks "can we say this?", "is this true?", "does the product really do X?", when a reviewer asks for a feature or a claim the product may not support, and when a source document (pitch deck, PRD, marketing page) makes claims the video wants to repeat.

josueh04/product-video-skills

Check a rendered product video before anyone else sees it: worker-pattern flicker, black frames, loudness and true peak, clipping, clicks at clip edges, overlapping narration, speech to text against the script, banned terms and legacy names, camera zoom, contact sheets, frame strips at transitions and parity against the approved version; then write qa/REPORT.json, the only thing deliver.py accepts. Use it after every HyperFrames render, whenever someone asks "is the render clean", "QA this", "check the video", "check the audio", "why does it flicker", "there is a click", "compare v3 with v2", "did the approved part change", or before showing, sending, uploading or delivering any MP4, even when the request does not say QA. Also use it to triage a defect a reviewer reported in a render.

josueh04/product-video-skills

Write a product video's narration and turn it into voice clips with word timings, pronunciation fixes, sound effects and even loudness. The script becomes a table of moments and then audio/lines.tsv (one clip per sentence, with role and speed columns); tts.py voices it with ElevenLabs or the free macOS say voice, maps brand respellings back to the on-screen spelling, normalizes every clip and writes audio/timings.json for the composer; make_sfx.py builds typing tracks from real keystrokes and places recorded click and pop sounds. Use it whenever a video needs a script, narration, voice-over, lines.tsv, timings.json, TTS, a new take, a voice or casting choice, a pronunciation fix ("it says the name wrong"), a changed sentence, a tone note ("too hype", "sounds cut off"), audio levels, a click at the end of a clip, typing or click sounds, or when the build stage asks for the voice. Also use it for silent loops, which still need a moment table and SFX.

josueh04/product-video-skills

The animation rules that keep HyperFrames' parallel render workers from dropping, flashing or flickering elements, plus a static lint (lint_motion.py) that finds the violations in a video's template before it costs a render. Use it whenever you write or edit GSAP tweens, timelines, cursors, camera moves, typing, scrolls or pop-ups in a HyperFrames composition or a video's src/template.tpl, whenever a render shows flicker, stutter, an element that vanishes on some frames, a title that flashes, or a "WORKER PATTERN" line from scan_render.py or qa.py, and whenever the preview looks right but the MP4 does not. Also use it to review someone else's timeline code before rendering.

josueh04/product-video-skills

Pin a product's source code (read-only exports in sources/ plus sources.lock), confirm that the pinned commit is what runs in production, and map every screen of a video brief to its route, components, i18n strings and state, written to the video's SOURCES.md. Use it whenever a video needs to know where a screen lives in the code, when sources/ is missing or stale, before ui-spec-from-code or product-truth start on a video, after the product's frontend changed ("what changed", "which videos are affected", "refresh the sources", "is this checkout current", "which commit is in prod"), and when there is no code and you need an inventory of the no-code references (recordings, recovered captures, docs) a screen can be rebuilt from.

josueh04/product-video-skills

Compose a narrated product demo in HyperFrames: the stage (the product UI rebuilt at its real viewport and scaled to 1080p, camera, rack focus with veil, chapter titles, cursor and clicks, typing, streaming text, pop-ups, toasts, scrolls, end screen and lockup) and the build.py that anchors every beat to a word of the narration. Use it whenever you write or edit a video's video/build.py, src/template.tpl or src/app.css, place a beat on a word, add a chapter, a click, a pop-up or a push-in, frame a screen, build the end screen or lockup, snapshot setup beats, or render a draft or delivery MP4 of a product video in this workbench. Also use it when someone says "the cursor is off", "too zoomed in", "too fast", "it feels chaotic", "the title flashes", "sync the UI to the voice", or asks for a walkthrough, demo or pitch video of a UI.

josueh04/product-video-skills

Gather pixel references for UI that the code cannot show, or to check a rebuild against the real thing, using frames and timed OCR text from screen recordings, captures recovered from past Claude Code session transcripts, a local instance of the app built like production, and web research for third-party apps, plus side-by-side parity images and contact sheets. Use it whenever someone hands over a screen recording (.mov or .mp4) of the product, when a screen has no usable source code (a stale checkout, another company's UI such as a sign-in page, calendar or CRM, runtime output from a backend not in the repos), when asked "what does it really look like", "match the recording", "how long does that animation take in the app", "compare our render to the real app", or when screenshots from an earlier session might already exist. Never uses the reviewer's personal browser.

josueh04/product-video-skills

Turn a product's real frontend code into 1:1 rebuild specs for a video, one read-only subagent per screen, each returning static HTML, CSS with every variable resolved to its literal value and cited (role/path:line@sha), every state, transitions with exact durations and easings, icons from the code's own icon sets, and the exact i18n strings; plus resolve_tokens.py to write the product's design tokens to kit/tokens.css. Use it whenever a screen of the product has to appear in a video, when writing or fixing video/src/app.css or the template markup, when someone asks for exact sizes, colors, fonts, paddings, animations or icons of a screen, when a rebuilt screen "looks off" next to the real app, and when the design tokens or theme of a product need extracting. Framework adapters cover Angular with PrimeNG (proven), React, Vue, Tailwind and plain HTML (unproven).

josueh04/product-video-skills

Say where this session stands in the Product Video Skills workbench (setup state, which product and video the current folder belongs to, the stage of every video) and the exact next command to type. Also answers "how do I..." questions about the workbench from its docs.

josueh04/product-video-skills

Start a new video of the current product - create videos/<video>/, write BRIEF.md, the feature coverage matrix (COVERAGE.md) and the claims sheet (CLAIMS.md), propose chapters, then stop for the reviewer's sign-off. Never builds. Run only when the user types /video-new.

josueh04/product-video-skills

Turn a batch of reviewer feedback on the product's videos into one table per video, fix every video that got notes in parallel (one subagent each) while keeping approved parts, re-run QA and parity, bump versions and deliver. Run only when the user types /video-review.

josueh04/product-video-skills

Check this machine and install the pinned video toolchain of the Product Video Skills workbench (HyperFrames CLI, its rendering Chrome and its agent skills from the same release, the Python environment, the speech model for QA), then run the self-check. Safe to run again. `/video-setup check` only reports.

相关技能

steipete/video-frames

Extract frames or short clips from videos using ffmpeg.

community

steipete/summarize

Summarize or transcribe URLs, YouTube/videos, podcasts, articles, transcripts, PDFs, and local files.

community

openclaw/openclaw

Use when testing, fixing, or extending the OpenClaw Control UI GUI with Vitest + Playwright end-to-end checks, mocked Gateway WebSocket flows, mocked dashboard runs, screenshots/videos, or agent-verifiable browser proof.

community

obra/superpowers

Obra superpowers agent skills.

community

finishing-a-development-branch

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work

community

writing-plans

Use when you have a spec or requirements for a multi-step task, before touching code

community