← ClaudeAtlas

breachlisted

Generate fast single-page HTML artifacts for daily development communication: status pages, HTML slide-like pages, flowcharts, PR writeups, incident pages, and structured discussion digests. Use the discussion-digest mode for email threads, GitHub issues/PRs, chat logs, and forum discussions when the user asks "梳理这个讨论", "这条线程结论是什么", "理一下参与人立场", "digest this thread", or "summarize this email chain". Use the general page mode for "quick HTML page", "快速 HTML 页面", "做个 HTML status page", "画一个 HTML flowchart", "做个 HTML slide page", "生成报告页面", "visualize this as a page", or "把这个做成 HTML". When the task already calls for a general page, support an optional ELI5 content profile for explicit requests such as "ELI5", "像给小孩解释", or "给新手讲". Not for native documents, native slide decks, spreadsheets, or full product visual design.
yuanhaorannnnnn/agent-skills · ★ 0 · AI & Automation · score 72
Install: claude install-skill yuanhaorannnnnn/agent-skills
# Quick Page Fast, single-page HTML artifacts for daily dev communication. ## Output Boundary breach owns HTML and deterministic discussion-digest rendering only. - Native document, presentation, or spreadsheet requested → use the requested OpenAI Template or native artifact capability. - HTML page, HTML slide-like page, or discussion digest requested → use breach. - Another workflow may define content and evidence first, then call breach only for HTML presentation. ## Core Rule Speed first, polish follows. A good-enough page in 30 seconds beats a perfect page in 5 minutes. Choose one mode: - **General page** — use html-effectiveness for layout and DESIGN.md sources for visual tokens. - **Discussion digest** — use the bundled schema, renderer, and template. Do not run the generic layout/style selection because this mode is deterministic. ## Optional Chart Renderer In General page mode, `lieflat-charts` may render a chart module inside the page. Breach still owns page integration and chart placement; `paperwork` owns the surrounding narrative. Use it only when all are true: - the input contains real quantitative data or an honest relationship graph; - a chart carries an independent conclusion better than prose, cards, or a table; - the visual encoding does not require invented scores, weights, or benchmarks; - the artifact's use is compatible with the renderer's license. Keep this route small: normally 1–3 charts, each with a distinct conclusion. Do not chart