Communitygithub.com

Animated Infographic

Create reproducible animated infographic GIFs from structured data or measured results, with audience-language labels, explicit timing semantics, representative-frame QA, and a static fallback.

What is Animated Infographic?

This skill enables the generation of animated infographic GIFs from structured data (JSON/CSV/text) or benchmark results, making it ideal for dynamic visualizations such as card movements, rankings, comparison bars, or state transitions. It distinguishes between illustrative animations (purely aesthetic), measured animations (based on actual timing data), and mixed modes, ensuring accuracy and traceability. Supported platforms include blogs, X/Twitter, and slides, with output tailored to the target audience's language and device dimensions (default 1200x675). The skill enforces a rigorous validation workflow: freezing the source artifact, designing a clear narrative, localizing all UI text, embedding explicit timing context in the asset, and performing both mechanical validation (via a Python script) and visual inspection of key frames. It is particularly useful for developers creating data-driven animations that must remain faithful to source measurements while being visually engaging for public or internal communication.

Works with~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/aktsmm/Agent-Skills/tree/HEAD/animated-infographic

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

What does Animated Infographic do?

When to Use

  • Create GIFs or animated graphics from JSON / CSV / benchmark results.
  • Show card movements, rankings, comparison bars, or state transitions with animation.
  • Create articles / SNS assets where numerical accuracy and the distinction between staging time and measured time are important.

For static PNGs only, use visualize-as-infographic; for analyzing existing videos, use video-watch; for editable schematic diagrams, use draw.io workflows.

Inputs

  • Source artifact: JSON / CSV / text. If measured values exist, this should be the single source of truth (SSOT).
  • Audience language and target medium: e.g., Japanese blog, X, slide.
  • Timing mode:
    • illustrative: Movement is for presentation only; does not represent processing speed.
    • measured: Frame timing is derived from actual measured timestamps.
    • mixed: Numerical values are measured; card/bar movement is for presentation.
  • Output size. Default is 1200x675 for blog/X; create a static fallback if necessary.

Workflow

  1. Freeze the evidence
  • Fix the source artifact, acquisition date, number of items, and measurement boundaries.
  • For latency, decide what is measured (e.g., 1 request, serial total, fresh-process end to end).
  • If processing external media, record the original URL, acquisition date, source relative path, and SHA-256 in a manifest; verify the hash before generation.
  1. Design the story
  • Limit one GIF to one claim. Return long explanations to the main text.
  • Fix the reading order (e.g., input → decision → next process, before → after, comparison → conclusion).
  • If a narrow decision gate is the subject, make it the visual main focus using accent color, thick outline, central placement, or a callout. Place input and output on either side, and write the value determined by the gate inside the box.
  • Do not assert official terms and the author's own layers/classifications with the same appearance. For custom models, indicate the scope within the asset (e.g., Organized in this article), and avoid arrows or nesting that implies product inclusion or maturity.
  1. Localize before rendering
  • Align explanatory UI, status, and captions to the reader's language. Do not leave generic English UI in Japanese articles.
  • Preserve official notation for proper nouns (product names, API fields, model IDs, etc.).
  • Avoid literal translations. E.g., use Which level of the defined criteria it is close to instead of Score at stage.
  1. Make timing explicit inside the asset
  • illustrative: Display Card movement is for presentation (does not represent actual speed).
  • measured: Display measurement boundaries and units; derive frame duration from actual values.
  • mixed: Write both Displayed values are measured / Bar/card movement is for presentation.
  1. Generate reproducibly
  • Keep the generator script, GIF, static fallback, and source snapshot if necessary.
  • Read values from the source artifact; do not enter them manually.
  • For third-party videos, limit to the minimum frames needed for the claim (e.g., before / operation / after), and exclude irrelevant browser chrome. Distinguish and display attribution and misinterpretation prevention information (e.g., dummy data, unsent, author-added notes) within the asset.
  1. Validate mechanically
python scripts/validate_animation.py output.gif `
  --expected-width 1200 --expected-height 675 `
  --timing-mode illustrative `
  --contact-sheet output-contact.jpg `
  --report output-validation.json
  1. Inspect visually
  • Check the start, middle, and end frames at actual size and target embed width (approx. 360px for mobile).
  • Fix and regenerate if issues are found: text overlap, clipping, ordering, whitespace, unreadable notes, or remaining English UI.

Timing Rules

  • The total playback time of the GIF does not necessarily match the API latency. Do not equate them automatically.
  • For the sum of multiple requests, display whether they are parallel or serial.
  • If measurement boundaries differ across comparisons, treat them as observed end-to-end time, not model speed.
  • For estimated costs, accuracy rates, and token counts, cross-check with the source artifact and include notes in the asset to prevent misinterpretation.

Done Criteria

  • GIF is animated and meets expected dimensions and minimum frame count.
  • Timing mode and measurement boundaries are visible within the asset.
  • Source artifact values match displayed values.
  • If from external media, source URL / acquisition date / SHA-256 match the manifest.
  • Explanatory UI is unified in the reader's language.
  • The subject decision gate is visible first, and what it receives and returns is readable.
  • Official definitions and the author's own organization are distinguishable within the asset alone.
  • Start, middle, end frames and loop playback have been verified.
  • Text is readable at target embed width.
  • Location of generator, GIF, static fallback, and validation report has been reported.

Individual skills in this repo

This repo contains 2 individual skills — each has its own dedicated page.

Related Skills