Communitygithub.com

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.

Was ist product-video-skills?

product-video-skills is a Claude Code agent skill that 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.

Funktioniert mit✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/josueh04/product-video-skills/tree/HEAD/skills/product-kit
Original anzeigen→Alle Skills durchsuchen

In Ihrer bevorzugten KI fragen

Öffnet einen neuen Chat, in dem dieser Agent-Skill bereits geladen ist.

OpenAIChatGPTClaudeClaudeGeminiGeminiGrokGrokPerplexityPerplexityDeepSeekDeepSeek

Dokumentation

Product kit

Every video of a product draws from the same kit, so the kit is extracted once, approved once, and reused. In the production these skills come from, doing this piecemeal was expensive: an icon font was downloaded without asking, then with approval, then widened three times, and in between a subagent took two glyphs from a font inside a browser extension folder; the same demo customer ended up with two different surnames in two videos because a list of names to veto was never answered; and a renamed product kept its old name on screen until a reviewer caught it.

Before you start

  • Sources are pinned (sources.lock, sources/<role>/). If not, run source-recon first: every kit file cites the locked commit.
  • Ask once for every download the kit needs (font packages, icon packages not in the repo), in a single list with what, where from and about how big. Nothing is downloaded without that yes, and nothing ever comes from a browser profile or extension folder.
PVS_HOME="$(cd "$(cd "${CLAUDE_SKILL_DIR}" && pwd -P)/../.." && pwd)"
K="$PVS_HOME/skills/product-kit/scripts"

1. Tokens

Run ui-spec-from-code's resolver for the product's theme, and write the render facts it describes (runtime rem, theme selector, fonts actually loaded) to kit/render-facts.md:

"$PVS_HOME/bin/pvs-py" "$PVS_HOME/skills/ui-spec-from-code/scripts/resolve_tokens.py" <product_dir> \
  [--theme-selector "<dark selector>"]

Set brand.accent in product.yaml from the token the product uses for its primary color, citing it.

2. Fonts

Find which fonts the app really loads (link tags, @font-face, font packages in the lockfile), not only the ones CSS names. Take the files from the repo or from the installed package (for example a @fontsource/<family> package in node_modules); a font that only exists on a web font service needs the user's approval to download. Then subset:

"$PVS_HOME/bin/pvs-py" "$K/subset_fonts.py" <product_dir> <font files...> \
  [--text-file videos/*/audio/lines.tsv] [--license OFL-1.1] [--source <citation>]

It writes kit/fonts/<family>-<weight>.woff2, rebuilds kit/fonts/fonts.css (font-display block, so a render never paints a fallback first), copies the license text it finds beside the font or in its package (the OFL requires the text to travel with the font) and records source, license and license file in kit/fonts/fonts.md. A font with an "unknown" license or a "missing" license file is not ready: find them or ask. Set brand.font in product.yaml. For an icon font that works by ligatures, pass --all-features and the ligature names as --text, and check each name renders as a glyph, not as letters.

3. Icons

Get the list of icons from ui-spec-from-code's icon extraction (or the specs), grouped by package. Copy them from the product's own packages:

"$PVS_HOME/bin/pvs-py" "$K/collect_icons.py" <product_dir> --package <dir with svg files> \
  --names calendar,check,plus [--variant outline/24] [--prefix bx-] [--clean]

It copies each SVG to kit/icons/<set>-<name>.svg and adds a row to kit/icons/icons.md (file, name, set, source citation, license). It fails on a missing or ambiguous name instead of picking a look-alike, because a near-miss icon is what made a reviewer reject a whole first cut. Icons that only exist as an icon font: subset the font (section 2) and take codepoints from the library's own CSS; the build must fail on an unknown name.

4. Logos

From the repo's assets (or a brand package the user hands over) only: never redrawn, traced or taken from a website. List the candidates, look at them, then copy the three variants:

"$PVS_HOME/bin/pvs-py" "$K/collect_logos.py" <product_dir>            # candidates in sources/
"$PVS_HOME/bin/pvs-py" "$K/collect_logos.py" <product_dir> --light frontend/assets/logo.svg \
  --dark frontend/assets/logo-white.svg --tile frontend/public/icon-512.png [--root <brand folder>]

It writes kit/logos/logo-light.* (for light backgrounds), logo-dark.* (for dark backgrounds) and app-tile.* (the square icon), SVG when the product has one, with a cited row each in kit/logos/logos.md; it refuses URLs. Then set brand.logo_light, brand.logo_dark and brand.app_tile in product.yaml to the paths it prints. Pick the variant by background in every video: a light mark on a white page disappears, which happened in the source production.

5. Cast

Demo data is fictional from the first frame, and it is the same in every video. Read references/cast-and-names.md for the reserved number ranges and domains. Propose one full cast in a single closed question, with your proposal as the recommended default:

Demo cast for every Acme Tasks video (recommended: use as is): Northwind Bakery, Portland OR (northwind.example); owner Maya Lindqvist, [email protected]; customer Omar Haddad, +1 503 555 0172. Use it, or change which names?

Full names, one surname per person, kept forever; numbers in 555-0100 to 555-0199; emails and sites on reserved domains. Reviewers veto names on taste (a bland name was rejected in the source production), so ask before anything is animated. If the reviewer does not answer, say you are freezing the proposal and freeze it: silence must not leave two versions of one person. Write it to cast: in product.yaml and check it:

"$PVS_HOME/bin/pvs-py" "$K/check_cast.py" <product_dir>

It fails on a missing surname, a first name with two surnames, a non-555 or non-fictional phone, and an email or website on a real-looking domain.

6. Names, pronunciations, banned terms

Ask the user, then write to product.yaml:

  • names: canonical name to the legacy strings that must never appear (old product names the live UI or the code may still show). Every video uses the canonical name even where the real UI is behind, and grep finds the legacy ones before a render.
  • pronounce: brand words to the spelling the voice engine says right (a made-up product name read as a different word was caught by ear in the source production).
  • banned_terms: competitors, third-party vendors the team keeps off screen, real customers, internal code names.
  • never_say and positioning: the one-line identity and the phrasings it must never take.

7. Read-only tools

Fill section 10 of kit/RULES.md, "Read-only tools" (the template ships it with a placeholder), with the product's API or MCP tools agents may call (docs, catalogs, templates), and say that everything else is off limits. A connection with write access stays read-only by this rule; recommend a credential scoped to read-only, because in the source production read-only was enforced only by instruction.

8. Report

Show the user the kit inventory (tokens count and flags, fonts with licenses, icons by set, logos, cast), what still needs approval, and the product.yaml keys you changed.

Individual skills in this repo

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

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

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.

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.

Verwandte Skills

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