agentleFS
stv-iostv-io/claude-skills/skills/diagram-style/SKILL.md

diagram-style

Skill0 starsChanged 3 days agoMITView on GitHub

House style for architecture and infrastructure diagrams — colour by role (open-color palette), Helvetica type scale, zone/card/chip/arrow rules, title-as-conclusion, legend, takeaway and provenance lines. Use BEFORE drawing or restyling any diagram on the Excalidraw canvas, in generated SVG/PNG figures, or when the user asks for diagrams that "look like the others", a consistent diagram style, or a style guide. Composes with excalidraw-skill (canvas mechanics) and aws-diagrams / gcp-diagrams (provider icons); provider conventions override this guide for their own figures.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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. Very specific:

    Read STYLE.md before drawing.

    line 8
  2. Very specific:

    Helvetica (fontFamily: 2) everywhere, nothing below 13: No bold exists in Excalidraw.

    line 19
  3. Distinctive:

    Look at the rendered PNG: before you call it done.

    line 27

≈ “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

Diagram style

Read STYLE.md before drawing. Take values from tokens.json rather than retyping hex codes.

The short version:

  1. Title = the conclusion, size 28. Subtitle 16 in secondary. Legend of only the roles and line styles used. Optional one-line takeaway; a provenance line ("Verified <date>: <how>") for anything showing live infrastructure.
  2. Colour by role, not vendor: managed blue, owned green, focus orange (one or two per figure), hub cyan, external pink, automation violet, risk red, outside grey, later grey dashed. Three or four roles per figure.
  3. Helvetica (fontFamily: 2) everywhere, nothing below 13. No bold exists in Excalidraw; use size and colour.
  4. Flat shapes: roughness: 0, solid fills, rounded corners, stroke 1 for cards and 2 for zones and the focus card. Zone titles are separate text at the top-left, never a bound label.
  5. Square-cornered arrows, filled triangle heads, coloured by the source's role; solid = now, dashed = later/optional/legacy, dotted = weak.
  6. Official provider icons (via the aws-diagrams / gcp-diagrams skills if you have them; separate, not in this repo); every icon has text.
  7. Look at the rendered PNG before you call it done.

Reference figure

Image: Reference figure

scripts/reference.py draws one figure that uses every role, line style and text size, straight from tokens.json. Use it to check the style renders as intended, or as a starting point for a new figure.

python3 scripts/reference.py --out reference.excalidraw          # write the file
python3 scripts/reference.py --post                              # add to the live canvas (does not clear it)
python3 scripts/reference.py --post --dy 1200                    # stack below existing work

The canvas URL defaults to http://127.0.0.1:3030; override with --canvas (the canvas server's own default is port 3000). Render a PNG with the Excalidraw MCP export_to_image tool.

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.