write-diagrams
Skill14 starsChanged 5 days agoMITView on GitHub
Use when a document produced in this collection (board, plan, report, analysis, durable doc) could carry a diagram, or when a design question would settle faster shown than said. Picks the diagram that answers the open question and holds the discipline that keeps a diagram readable. Not for decoration: no open question, no diagram.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Ovich/swarmstormword for word, first seen Oct 25, 2026
Other versions1 project uses a version of this.
Word for word · 1
- Ovich/swarmstorm★ 14100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- brainstormSkill
- brainswarmSkill
- interviewSkill
- whiteboard-defenseSkill
- complexity-reworkSkill
- deepen-moduleSkill
- dependency-graphSkill
- simplify-reworkSkill
- simplifySkill
- aiviewSkill
- swarm-boardSkill
- execute-planSkill
- long-horizonSkill
- orchestrate-planSkill
- guide-complexitySkill
- guide-createSkill
- guide-deepen-moduleSkill
- helpSkill
- next-stepSkill
- orchestrateSkill
- cold-readSkill
- write-planSkill
- roadmapSkill
- technical-writingSkill
- write-conventionsSkill
- pr-reviewSkill
- recurring-review-findingsSkill
- swarmSkill
- design-prototypeSkill
- point-atSkill
- validate-deliverySkill
- verification-skill-createSkill
- create-workbenchSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
A missing diagram is a finding, not a pass: when a downstream document depends on a boundary no diagram shows, draw it then.
≈ “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.
Files
Write diagrams
A diagram makes a boundary, an ordering, or a dependency visible early enough to argue about cheaply, and stays as compressed context for the next reader.
The catalog
diagrams.md maps each open question to the diagram that answers it. Read it before drawing. No open question, no diagram.
Draw while a question is open, not only in the final document: the container diagram with a ? on the contested arrow, and ask "which of these two?".
The discipline
- One diagram per open question. A diagram that restates a paragraph is deleted.
- Every diagram carries a one-line caption right above the fence: its catalog type and the question it answers, "Dependency graph: what may import what after this change."
- Draw only what one sentence cannot say.
- Every arrow is labeled with what flows and in which direction, unless the shape carries it: when every arrow means the same thing and no node has two outgoing arrows, the node names already say it. The moment a node branches, or two arrows mean different things, every arrow gets a label.
- Real names of real things: actual service, table and module names, never
Service A. - A box whose responsibility you cannot state in a phrase is deleted.
- 2–4 diagrams for a feature. The full C4 set only for a genuinely new system.
- When the design changes in review, update the diagram in the same edit.
- A missing diagram is a finding, not a pass: when a downstream document depends on a boundary no diagram shows, draw it then.
Mechanics
- Mermaid in fenced ```mermaid blocks, so it renders in aiview and on any git host.
- C4-style diagrams as
flowchartwith C4 conventions (classDeffor person / system / external); Mermaid's nativeC4Contextblocks render inconsistently. - Colours as eight-digit hex, never
rgba(...); the commas insidergba()split aclassDefand the definition fails silently. - Translucent over opaque (
#7f7f7f1a); a translucent fill works in both themes, an opaque light fill burns a white slab into a dark page. - Meaning carried by styling is in the label too, a glyph or a word, so it survives grayscale, dark mode and a colour-blind reader.
- Look at the diagram in aiview after every edit. A block that fails to parse shows
mermaid failedand its source in place.
Red flags
| Thought | Reality |
| "The prose already covers the boundary" | A boundary described in a bullet is a boundary nobody checked. "What may import what" is a dependency graph, not a sentence. |
| "I'll invent a hybrid diagram" | The catalog's forms exist because each answers one question. Pick one; two questions get two diagrams. |
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
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.