SKILL

visual-portfolio

Divergent orchestrator: given one problem, produce a small portfolio of genuinely-different visualizations — each answering a DIFFERENT analytical question about the same data, each individually valid, each shipped with a trade-off card. Use when the ask is "show me the alternatives / what are the ways to look at this / a few views of X", or whenever a single figure would hide legitimate competing readings. Runs generate → validate → select → characterize. It sits ON TOP OF visual-critic (which validates and scores each member); it does not replace or duplicate the single-figure machinery.

原文 · 译文尚未补齐

适用对象:对象待分类。自动收录 · 待核查。

原作者说明节选

从固定版本的 SKILL.md 提取,保留原文语言。以下内容属于作者说明,本站尚未验证安装、兼容或运行结果。

查看英文原文

原文涵盖的主题

  • Visual Portfolio (divergent orchestrator)
  • Purpose
  • When to invoke
  • Inputs
  • Required intermediate representation
  • Procedure — generate → validate → select → characterize
  • Graceful degradation (honesty over quota)
  • The trade-off card (per member)

Turn one problem into a Pareto set of distinct, valid views rather than one "best" figure. Where visual-critic converges to the single strongest honest figure, this skill diverges: it enumerates the analytical questions the data can honestly answer, builds the best figure for each, keeps only the valid ones, and selects the few that together give the reader the most complete, non-redundant understanding of the stated decision. Each member ships a trade-off card, so three pictures become three arguments. The single failure mode this skill exists to prevent: variety without contribution — three chart-form swaps of the same question masquerading as a portfolio. Diversity here is defined over analytical questions (principles/analyticalquestions.yaml), never over chart types. The user asks for alternatives, options, "a few ways to see this," or an exploration rather than a single verdict. A single figure would suppress a legitimate competing reading (multiple decision-relevant questions live in one dataset). Building an exploratory panel, a "choose-your-view" page, or a section that must serve more than one audience/question at once. NOT for a single-figure deliverable with one clear qu

这里只展示节选,请以完整原文和当前要求为准。

来源仓库
Lmandrake/MandrakeWorkflows
源文件
skills/visual-portfolio/SKILL.md
固定版本
1874105936272447e2f9c81ba1dac48dae6ed074
阅读完整原作者说明 ↗

准备条件

  • 本条仅收录原作者的技能元数据;请阅读完整说明,确认依赖、许可与使用条件。

第一件可以尝试的小事

先阅读原作者说明,再决定是否在自己的 AI 工具中使用。

对应 AI 工具说明

作者有说明不等于本站实机验证。没有证据时,不自动标为兼容、安全或免费。

来源与内容状态

本条为自动发现的元数据,尚未完成人工文档核查;适用性与运行结果待核查。

原作者:Lmandrake/MandrakeWorkflows · 原作者仓库元数据。文档核查:未核查。最近来源检查:2026-11-11T03:01:44.755Z。

分类由名称、说明与路径规则生成,分类不代表已核实适配。

许可与权利归属

MIT。第三方内容及版权归原作者,使用条件以原文为准。

许可原文 ↗

本站没有因同步、复制或用户自报成功而将资源标为运行验证通过。

限制与注意事项

  • 元数据收录不代表本站已经安装、适配、实测或进行安全审查。