Help us improve
Share bugs, ideas, or general feedback.
From playwright-recast
Generates benefit-driven voiceover scripts and SRT subtitles from technical video descriptions or Gherkin feature files. Transforms UI actions into marketing narration for demo videos.
npx claudepluginhub thepatriczek/playwright-recast --plugin playwright-recastHow this skill is triggered — by the user, by Claude, or both
Slash command
/playwright-recast:script-writerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Transform technical descriptions of video content into polished voiceover scripts and subtitle text.
Generates video scripts for product demos, feature walkthroughs, launch videos, and social media from marketing briefs, blog posts, PRs, changelogs, git ranges, codebases, or freeform text.
Writes casual Hebrew narration scripts for video tutorials in Aviz's friendly style, using analogies, screen pointing, and proof endings. Turns video analysis into scripts before speech generation.
Generates timed, screen-by-screen walkthrough video scripts for app features with narration, screen directions, transitions, and shot lists using interactive agent workflow.
Share bugs, ideas, or general feedback.
Transform technical descriptions of video content into polished voiceover scripts and subtitle text.
.feature fileThe script can be generated from:
playwright-recast + playwright-bdd uses Gherkin .feature files where doc strings are the voiceover/subtitle text:
Feature: Product demo
Scenario: Setup monitoring
When the user opens the chat
"""
This doc string becomes the voiceover and subtitle for this step.
"""
And the user selects a skill
"""
Each step gets its own doc string — one per video segment.
"""
How it works:
Given/When/Then/And) = one video segment"""...""") under each step = voiceover + subtitle for that segmentSubtitleEntry[] via .subtitlesFromTrace()the user opens the chat) is the technical actionWhen writing doc strings:
Never describe the UI mechanically. Translate every action into client value — what it saves, simplifies, speeds up, or improves.
Every script follows this 4-part arc:
Adapt output to the input source:
From free-form description:
VOICEOVER:
[Continuous flowing text]
SUBTITLES:
[Same content split into short lines/blocks]
From Gherkin feature file — produce the updated feature with doc strings:
When the user opens the chat
"""
[narrator text for this step]
"""
SRT format (when requested):
1
00:00:00,000 --> 00:00:06,000
[subtitle text for segment 1]
2
00:00:06,001 --> 00:00:12,000
[subtitle text for segment 2]