agentleFS
Sign inStart free
kimgeakimgea/agent-kit/skills/build-interactive-diagram/SKILL.md

build-interactive-diagram

Skill0 starsChanged 32 days agoMITView on GitHub

Create polished, self-contained HTML diagrams for architecture, workflows, state transitions, comparisons, timelines, and other explanations that materially benefit from interaction. Use when a browser visual would communicate relationships better than terminal prose, especially in remote Codex or Claude conversations; optionally publish the result through the serve-artifacts skill.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a temporary browser artifact when interaction materially improves the explanation.

    line 8
  2. Very specific:

    Create a new bounded output directory outside the installed skill.

    line 16
  3. Distinctive:

    Keep everything local: HTML, CSS, JavaScript, SVG, and data files.

    line 20
  4. Distinctive:

    Read diagram-quality.md while building or reviewing the visual.

    line 25
  5. Distinctive:

    Check the browser console and ensure all assets resolve without network access.

    line 33
  6. Very specific:

    If serve-artifacts is available, publish the directory with a short relevant TTL and return its browser URL.

    line 34
  7. Distinctive:

    If the host is unavailable, return the absolute entry-file path and concise local serving instructions.

    line 36
  8. Distinctive:

    Do not embed private context, credentials, raw transcripts, or unrelated repository files.

    line 41

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Build Interactive Diagram

Create a temporary browser artifact when interaction materially improves the explanation. Use prose or a small Markdown table for simple facts and one-step flows. Keep long-term product documentation in its owning repository.

Build the visual

  1. Choose the smallest useful form: flow for sequence, graph for dependencies, timeline for change, or grouped cards for comparison and ownership.
  2. Create a new bounded output directory outside the installed skill. Copy assets/starter/ into it when the starter matches; never edit the installed copy.
  3. Replace the sample model and labels with the user's actual content. Remove controls that do not help the explanation.
  4. Keep everything local: HTML, CSS, JavaScript, SVG, and data files. Do not add a CDN, analytics, external fonts, or a framework merely for presentation.
  5. Make the first viewport understandable without interaction, then use selection, filtering, focus, or progressive detail to reveal more.

Read diagram-quality.md while building or reviewing the visual. It contains the layout, interaction, accessibility, and QA contract.

Validate and deliver

  • Open or request the entry page through a local HTTP server; do not claim visual quality from source inspection alone.
  • Exercise every control, keyboard focus, narrow layout, empty state, and long label.
  • Check the browser console and ensure all assets resolve without network access.
  • If serve-artifacts is available, publish the directory with a short relevant TTL and return its browser URL. Follow that skill for remote/Tailscale setup.
  • If the host is unavailable, return the absolute entry-file path and concise local serving instructions. The diagram skill must remain useful on its own.
  • Summarize what the visual shows in the conversation so the result is still useful to someone who cannot open it.

Do not embed private context, credentials, raw transcripts, or unrelated repository files. When revising an existing artifact, create a new published copy and revoke the old one after the user confirms it is no longer needed.

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.

Check your setup, free