一键导入
scopibuild
Capture screenshots and generate PNG/PDF output from existing HTML slides
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Capture screenshots and generate PNG/PDF output from existing HTML slides
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Full card news generation pipeline — auto-selects sequential (subagent) or collaborative (Agent Teams) mode based on environment
Show all Scopi Card News commands, agents, and documentation
Agent Teams reference — detailed inter-agent communication protocols and task architecture (auto-invoked by /scopi:generate when Agent Teams is available)
Brand identity interview — deep setup wizard that captures your voice, audience, visual identity, and generates a unique theme
Social media caption and hashtag generation with DARI and BINNA — platform-optimized captions for your card news
Content ideation with NARA using VS (Verbalized Sampling) — brainstorm topics, story arcs, and series with creative alternatives
| name | scopi:build |
| description | Capture screenshots and generate PNG/PDF output from existing HTML slides |
| user_invocable | true |
You are running the Scopi build pipeline. This takes existing HTML slides and runs the Puppeteer capture + PDF assembly pipeline. It can also run Playwright screenshot captures before building.
Read scopi.config.json for dimensions, pipeline settings, and capture targets.
The user can specify:
/scopi:build output/my-topicoutput/ for the most recent subdirectory with .html files, or falls back to output/html/)If pipeline.capture is true and capture targets exist (either in config or in assets/captures/):
Check if captures need to be refreshed. If assets/captures/ is empty or user requests fresh captures:
const { captureAll } = require('./templates/capture.js');
const config = require('./scopi.config.json');
const captures = await captureAll(
(config.identity.captureTargets || []).map(url => ({
name: url.replace(/https?:\/\//, '').replace(/[./]/g, '-'),
url: url,
viewport: `${config.dimensions.width}x${Math.round(config.dimensions.width * 0.75)}`
})),
{ outDir: 'assets/captures' }
);
Check the specified directory for .html files. If no directory specified, scan output/ for the most recent subdirectory containing .html files. If none found, report the error and suggest running /scopi:generate first.
Execute the generation pipeline:
node templates/generate.js --html=[html-dir] --out=[output-dir]
Build Complete!
Output: [output-dir]/
[N] PNG slides
carousel.pdf
Captures: [N] screenshots (if any)
Open [output-dir]/ to preview.
--retina / --no-retina — Override 2x scaling--format=png — Skip PDF generation--format=pdf — Skip PNG output (PDF only)--width=N --height=N — Override dimensions--capture — Force fresh screenshot captures before building