Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Diagramming

ASecurity

Draw BA diagrams in mermaid — sequence, flow, state, erd — derived from the entity spine first, free text second, and ship every one compiled or labelled [UNRENDERED], never a third state. Use when a PRD/SRS needs a picture of who-calls-whom, a decision flow, an entity's lifecycle, or the data model, or before /ba:spec compose embeds one.

3 stars
0 votes
0 copies
0 views
Added 9/23/2026
documentationgonodeawsgitdocumentation

Works with

cli

Security Analysis

A100/100

Scanned 9/23/2026

Install to Claude Code

$npx -y skills add trungdo9/ClauKit --skill diagramming --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Diagramming?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Diagramming
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/trungdo9-diagramming/badge)](https://www.skillsdirectory.com/skills/trungdo9-diagramming)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: diagramming
description: Draw BA diagrams in mermaid — sequence, flow, state, erd — derived from the entity spine first, free text second, and ship every one compiled or labelled [UNRENDERED], never a third state. Use when a PRD/SRS needs a picture of who-calls-whom, a decision flow, an entity's lifecycle, or the data model, or before /ba:spec compose embeds one.
license: MIT
---

# Diagramming

One command (`/ba:diagram`), four mermaid types, one render gate. Wave 0 — D-10 cut
D2/PlantUML/BPMN/DBML and three of the seven mermaid diagram kinds (`context`/`dfd`/`journey`/
`class` wait for a real project to ask; see the capability map).

## The four types, and what each is for

| type | answers | draws from |
|---|---|---|
| `sequence` | who calls whom, in order | an `FR`'s `Actor:` line + its `UC` children's steps |
| `flow` | activity/decision; `subgraph` blocks stand in for swimlanes | a `UC`'s numbered steps |
| `state` | one entity's lifecycle | the `FR`/`AC` pair that names a transition (`HOLD → BOOKED` is FR-011/FR-012's shape) |
| `erd` | the data model | the nouns the spine implies, not the spine's own node kinds |

A type that does not answer a question the spec is already asking should not be drawn — a `state`
diagram for an entity that never changes state is decoration, not documentation.

## Input is the spine first, free text second

Given `FR-012`, read the entity and its `UC`/`US` children — read the `traceability` skill file
first (link below) — and draw from them, so the diagram and the spec cannot disagree. Free text is
the fallback for a whiteboard moment before an FR exists, and what it produces carries
`confidence: low` like anything else unsourced (rule 2) — a diagram is a claim about the system
the same as any entity, and an unsourced claim is labelled, never presented as settled.

## Why mermaid only

Claude Artifacts render mermaid natively with no binary; GitHub renders it inline in markdown; and
this machine has no renderer installed for anything else (`d2`, `plantuml`, `dot` all absent,
alongside `mmdc`). One notation that always displays beats five that display sometimes. Other
notations arrive when a real project asks (YAGNI, D-10 made it explicit) — capability map row 2
(`erd` in D2/dbdiagram) is filed, not built.

## The render gate (rule 4), and how it degrades — the whole of its teeth

Try `npx -y @mermaid-js/mermaid-cli` (`mmdc`) first. Two outcomes, and only two:

- **Renderer reachable, block compiles** — ship the compiled artifact; no `[UNRENDERED]` label
  anywhere in the file. A block that fails to compile is fixed before it ships, not shipped with
  an apology.
- **Renderer unreachable** (this machine, today — `mmdc` not found, and it may be offline besides)
  — syntax-check what can be checked locally, ship the mermaid source, and open the file with the
  `[UNRENDERED]` header block (`references/mermaid-patterns.md`) verbatim. Never present an
  unverified diagram as verified — the same posture `seo-drift` takes with `[NO BASELINE]` rather
  than inventing one.

No third state. A file that is neither compiled nor labelled fails the exit gate — that third
state is exactly what rule 4 exists to forbid.

## Anti-patterns (auto-reject)

- A 40-node `flow` nobody can read — split it, or state in prose what the picture cannot carry.
- A `sequence` diagram with no actor from `plans/ba-context.md` § 2 — an unnamed stick figure is a
  placeholder, not a diagram.
- Re-drawing in mermaid exactly what the composed SRS already says in prose — a diagram earns its
  place by answering a question prose answers badly (order, branching, state), not by repeating it.

## Reading order

- `references/mermaid-patterns.md` — one minimal known-good block per type, the diacritic-quoting
  rule, and the `[UNRENDERED]` header block verbatim.

## Cross-references

- **Read the `traceability` skill file** ([.claude/skills/ba/traceability/SKILL.md](../traceability/SKILL.md)) — the entity contract this skill reads before drawing anything.
- **Read the `business-analysis-rules` workflow** ([.claude/workflows/business-analysis-rules.md](../../../workflows/business-analysis-rules.md)) — rule 2 (confidence, for free-text input), rule 4 (the render gate this skill implements), rule 6 (hard-fail pre-flight), rule 7 (output language).
- `.claude/commands/ba/diagram.md` ([../../../commands/ba/diagram.md](../../../commands/ba/diagram.md)) — the four actions this skill backs.

Attribution

trungdo9trungdo9
View sourceMore from trungdo9 →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Context Fundamentals

Understand the components, mechanics, and constraints of context in agent systems. Use when designing agent architectures, debugging context-related failures, or optimizing context usage.

179001 votes

release-notes

Draft release notes and changelog entries from git history or merged PRs between two refs (tags/SHAs/branches), including breaking changes, migrations, and upgrade steps. Use when the user asks for release notes, changelog updates, or a GitHub Release draft.

1301 votes

docs-style-guide

Documentation style guide enforcer by @planetabhi. Applies and reviews the writing style guide when authoring or editing product documentation and tutorials. Use to check prose for voice, tense, word choice, inclusive language, formatting, code block, UI, Markdown, and number/date conventions.

11 votes

Caveman Help

Quick-reference card for caveman modes, skills and commands. Trigger: /caveman-help or "caveman help".

1066600 votes

How It Works

Explain how claude-mem captures observations, when memory injection kicks in, and where data lives. Use when the user asks "how does claude-mem work?" or "what is this thing doing?".

942310 votes
View all in documentation →