agentleFS
magnus919magnus919/agent-skills/infographic/SKILL.md

infographic

Skill172 starsChanged 4 months agoMITView on GitHub

Design, create, or improve an infographic, visual explainer, illustrated guide, comparison poster, process graphic, or data story for any audience. Use for turning evidence and ideas into a clear, polished visual explanation, including choosing information structure, visual encoding, and a distinct aesthetic. Do not use for standalone decorative artwork, complete brand identity systems, statistical analysis alone, or ordinary document formatting.

13 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use brand-designer for identity systems, data-scientist for statistical inference, documents for office/PDF production, and color-management

    line 22
  2. Distinctive:

    Use the brief for complex work.

    line 39
  3. Distinctive:

    Ground the message: Read evidence and editorial judgment for factual, quantitative, persuasive, or source-poor requests.

    line 40
  4. Distinctive:

    Choose the information structure: Read layouts and encodings when composing or revising.

    line 47
  5. Distinctive:

    Choose a visual language separately: Read style recipes when a style is requested or needs choosing.

    line 52
  6. Distinctive:

    Execute with real capabilities: Read production before rendering.

    line 58
  7. Distinctive:

    Inspect and improve the actual export: Read review and polish before delivery.

    line 65
  8. Distinctive:

    Stop when the artifact meets its communication contract and review gates, or when a specific missing capability/source prevents completion.

    line 74

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Infographic

Produce a visual explanation that helps its intended reader understand, compare, choose, or act. A beautiful image with incorrect relationships or unreadable content is unfinished. This skill supplies judgment and design specifications; it does not install a renderer or confer access to an image-generation service.

When not to use

Use brand-designer for identity systems, data-scientist for statistical inference, documents for office/PDF production, and color-management for ICC/profile operations. These optional companion skills can support an infographic without replacing its editorial decisions. If a companion is not installed, use available tool guidance; do not assume a sibling path or capability exists. For decorative artwork alone, use the available image tool's own workflow.

Workflow

  1. Frame the communication contract. Establish who will read it, what they already know, and what they should understand or do afterward. Identify the real viewing context, dimensions, language, accessibility needs, and requested deliverables. Infer low-risk defaults and state them briefly; ask only for missing information that materially changes correctness or usefulness. Scale the process to the request: a small family guide does not need a long brief. Use the brief for complex work.
  2. Ground the message. Read evidence and editorial judgment for factual, quantitative, persuasive, or source-poor requests. Write the principal takeaway and the few supporting claims before decorating. Preserve material limitations beside affected claims. Do not invent statistics, citations, or causal relationships. A request for impressive numbers does not supply evidence. If sources are missing, research when available or produce a clearly qualitative/illustrative version and explain the limitation.
  3. Choose the information structure. Read layouts and encodings when composing or revising. Match the reader's task and real relationships to a comparison, sequence, timeline, mechanism, hierarchy, geography, data story, or reference arrangement. Sketch the reading order and focal point. Arrows, areas, containers, proximity, and numbering all make claims; use them deliberately.
  4. Choose a visual language separately. Read style recipes when a style is requested or needs choosing. Select a coherent direction appropriate to the reader, subject, medium, and supplied brand constraints. Offer a few genuinely different directions only when the choice is consequential; otherwise choose and explain one. A timeline can be a spare editorial graphic, a playful cut-paper story, or a technical plate without changing chronology.
  5. Execute with real capabilities. Read production before rendering. Check available tools, fonts, export formats, and inspection routes. Use deterministic geometry and typesetting for exact numerical charts and dense required copy. Use image generation for expressive conceptual imagery where available; hybrid assembly is valid. Follow the active image tool's rules. A text prompt is a production specification, not a rendered artifact. Do not claim execution or file formats that were not produced.
  6. Inspect and improve the actual export. Read review and polish before delivery. Verify evidence, geometry, visual semantics, hierarchy, actual-size legibility, accessibility, and export integrity separately. Repair content and structure before cosmetic details. Re-render after consequential edits; check the final file, not only the source or previous preview.
  7. Deliver what exists. Provide the requested visual, editable source when available, an accessible text equivalent, and traceable source notes. Briefly state meaningful limitations and what was tested. If rendering is unavailable, deliver a usable specification and exact copy, explicitly marked unrendered. Stop when the artifact meets its communication contract and review gates, or when a specific missing capability/source prevents completion. After two non-improving repair attempts on the same defect, change the approach or report the blocker rather than cycling indefinitely.

Resources on demand

  • Composition specification: for tool handoffs, hybrids, or layouts with multiple information layers.
  • Generation prompt: when using an image backend; adapt to its actual capabilities and keep verified text/data separately.
  • Sources and reuse: when checking methodological claims, citing foundations, or selecting outside assets.

Local artifact creation is the default. Publishing, sending, or overwriting an external artifact changes external state: confirm target, scope, and rollback path before the first mutation, using authorization already given in the conversation.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.