gooseworks-ai/goose-skills / skills/video/render-chatgpt-chat
render-chatgpt-chat
VerifiedAssemble a ChatGPT chat-reveal video ad from a thread + timeline JSON — one continuous Playwright recording of a ChatGPT mobile chat (user types with the iOS keyboard up → taps send → keyboard slides down + header cluster swaps in one beat → one gray loading dot → the assistant answer streams in word-by-word) crossfaded into a designed end card, with subliminal ChatGPT SFX and an optional ducked music bed. FREE assembly (Playwright + ffmpeg); the recipe supplies the per-brand thread + timeline + end-card config and gates the paid music call to its own capability. The ChatGPT sibling of render-imessage-chat. Use for the chatgpt-chat format.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.3 KB |
| assets/sfx/key-tap.wav | binary | 3.1 KB |
| assets/sfx/response-done.wav | binary | 13.9 KB |
| assets/sfx/send-tap.wav | binary | 7.8 KB |
| assets/sfx/stream-tick.wav | binary | 2.0 KB |
| scripts/config.example.json | config | 3.5 KB |
| scripts/end-card.template.html | other | 2.5 KB |
| scripts/mockup/generate.js | script | 16.3 KB |
| scripts/mockup/templates/chat.css | other | 16.6 KB |
| scripts/mockup/templates/icons.js | script | 11.5 KB |
| scripts/package.json | manifest | 320 B |
| scripts/record-chat.jsexecutable | script | 14.5 KB |
| scripts/render-end-card.jsexecutable | script | 6.2 KB |
| scripts/stitch.shexecutable | script | 6.6 KB |
| skill.meta.json | config | 229 B |
| tests/smoke-test.md | markdown | 2.4 KB |
tests/smoke-test.md
View on GitHub# Smoke test — render-chatgpt-chat
Verifies the free assembly end-to-end from the bundled example config. No paid
calls. Needs: Node, `npm install` (Playwright Chromium), and ffmpeg/ffprobe.
## Setup
```bash
cd scripts
npm install
npx playwright install chromium
mkdir -p /tmp/gpt-chat-smoke
```
## Run
```bash
node record-chat.js --config config.example.json --out-dir /tmp/gpt-chat-smoke
node render-end-card.js --config config.example.json --out-dir /tmp/gpt-chat-smoke
bash stitch.sh \
--chat /tmp/gpt-chat-smoke/master-chat.mp4 \
--end /tmp/gpt-chat-smoke/scene-end-endcard.mp4 \
--sfx /tmp/gpt-chat-smoke/master-chat.sfx.json \
--out /tmp/gpt-chat-smoke/master-final.mp4 \
--pad-color "#ffffff" \
--also-1x1
```
## Expect
- `master-chat.mp4` — 750×1624 (~9:19.5), ~12s, ONE continuous take: the user
types with the keyboard up, taps send (bubble + keyboard-down + header-swap in one
beat), one gray loading dot for ~500ms, then the assistant answer streams in
word-by-word. No micro-flicker / scene cuts.
- `master-chat.sfx.json` — `{ "cues": [ … ] }` with key-tap / send-tap / stream-tick
/ response-done entries and **no cue on the loading dot**.
- `scene-end-endcard.mp4` — 1080×1920, ~2.5s designed end card (wordmark + ⭐ proof
row + trust trio + CTA pill).
- `master-final.mp4` — chat crossfades into the end card (padded to 750×1624 with the
card bg so the seam is invisible); audio carries the subliminal SFX (and the ducked
bed if `--music` was passed). `master-final-1x1.mp4` is a 750×750 center crop.
- `ffprobe` confirms dimensions/duration; run the `watch` skill on the master to
confirm beat order, the single loading dot, and word-by-word streaming.
## Fail signals
- Chat looks horizontally fat/stretched → an export forced a non-9:19.5 ratio onto
the 750×1624 master. Never scale the chat to a different ratio (the 1:1 crop is the
only legitimate shape change).
- Assistant text appears all at once → the assistant message is missing
`"stream": true` (the mockup wraps words only when streaming).
- Keyboard visible while the answer streams → a `keyboard-hide` event is missing at
the send-tap `t`.
- Three dots instead of one → wrong row type; use a single `loading-dot` message.
- Visible seam under the end card → `--pad-color` didn't match `end_card.bg`.
- `Cannot find module 'playwright'` → run `npm install` in `scripts/`.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |