Generated banner for Markdown Mermaid Writing
Generated logo for Markdown Mermaid Writing

Markdown Mermaid Writing

K-DenseCommunityCreated by Clayton Young / Superior Byte Works, LLC (@borealBytes)Distributed by K-Dense

  • Skill
  • Tool integration
  • Documents and content
  • Apache-2.0
  • Low risk

From the author

Writes scientific Markdown documentation and Mermaid diagrams for workflows, relationships, timelines, and schemas. Provides syntax references, document templates, accessibility guidance, and version-aware rendering checks. Use when a user requests Markdown, Mermaid, or a text-based structural diagram; quantitative scientific figures require suitable plotting tools.
— the description by Clayton Young / Superior Byte Works, LLC (@borealBytes)

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill provides a structured approach to creating technical and scientific documentation using Markdown and Mermaid. It emphasizes the use of text-based diagrams for structural clarity, ensuring that workflows, architectures, and data models remain editable, version-controlled, and accessible. The skill includes templates for various document types and comprehensive guides for 23 different Mermaid diagram types, ensuring consistent formatting and adherence to accessibility standards.

What it does

  • Writes scientific Markdown reports, READMEs, and research documentation.
  • Creates structural Mermaid diagrams for workflows, timelines, schemas, and architectures.
  • Provides templates for common technical documents like ADRs, pull requests, and status reports.
  • Offers style guides for consistent Markdown formatting and Mermaid diagram design.
  • Includes accessibility guidance for diagrams using titles and descriptions.
  • Supports version-aware rendering checks for Mermaid diagrams.

When to use it

  • When you need to document complex workflows, system architectures, or data relationships.
  • When you want to maintain diagrams as editable text within a Git repository.
  • When creating technical reports that require a consistent, professional house style.
  • When you need to ensure your diagrams are accessible and renderable across different platforms.

Advantages

  • Diagrams are stored as plain text, making them easily diffable and version-controlled.
  • Reduces token usage compared to long-form text descriptions of relationships.
  • Promotes a consistent documentation style across projects.
  • Supports accessibility through standardized metadata.
  • Facilitates easy updates to diagrams without needing specialized image editing software.

Usage examples

“Create a Mermaid flowchart for a data processing pipeline.”
Generates a Markdown-formatted document containing a Mermaid flowchart block that maps the data pipeline steps, following the established style guide.
“Draft an Architecture Decision Record (ADR) for a new database choice.”
Produces a structured Markdown document using the ADR template, including placeholders for context, decision, and consequences.
“How do I create a sequence diagram for an API interaction?”
Provides the specific Mermaid sequence diagram syntax, a copy-paste template, and tips on how to style it according to the project's guidelines.
“Check if my Mermaid diagram is accessible.”
Reviews the diagram code for required accessibility metadata (accTitle/accDescr) and suggests improvements based on the Mermaid style guide.

Install

  1. Download the skill as a ZIP:

    https://mp-staging.autogineer.com/agent-skills/k-dense-ai/markdown-mermaid-writing/download
  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip markdown-mermaid-writing.zip -d .agents/skills

Documentation by Clayton Young / Superior Byte Works, LLC (@borealBytes)

Markdown and Mermaid Writing

Overview

This skill provides an optional house style for Markdown documentation with Mermaid structural diagrams. Follow the requested output format, existing repository conventions, and journal requirements first. This skill does not replace scientific analysis or plotting.

A relationship expressed as Mermaid inside a .md file is editable text that diffs cleanly in git. A compatible host renders it without a separate user build step. It renders where the host has Mermaid support; plain Markdown viewers may show only code, and platforms/extensions ship different Mermaid versions. It uses a compact source representation, though token use depends on the diagram. Exporting SVG/PNG requires a renderer; retain the diagram source alongside exported figures.

"The more you get your reports and files in .md in just regular text, which mermaid is as well as being a simple 'script language'. This just helps with any downstream rendering and especially AI generated images (using mermaid instead of just long form text to describe relationships < tokens). Additionally mermaid can render along with markdown for easy use almost anywhere by humans or AI."

— Clayton Young (@borealBytes), K-Dense Discord, 2026-02-19

When to Use This Skill

Use this skill when:

  • Writing a requested Markdown report, README, methods overview, or research note

  • Diagramming a workflow, data pipeline, schema, state machine, or conceptual relationship

  • Maintaining editable Mermaid sources and their rendered SVG/PNG exports

Use scientific plotting tools directly for measured data, uncertainty, statistical graphics, exact geometry, or publication figures that Mermaid cannot faithfully represent. Mermaid is not a required precursor to a quantitative chart or an existing SVG/code-native asset.

🎨 The Source Format Philosophy

Why text-based diagrams win
What mattersMermaid in MarkdownPython / AI Image
Git diff readable✅ text sourcePlotting code/SVG can also be text
Editable source✅Plotting scripts and vector sources are editable
Compact relationship notationOftenDepends on representation
Native previewHost/version dependentViewer/format dependent
Parseable by AI without vision✅❌
Works in destinationCheck Mermaid support and versionCheck image format support
Accessible (screen readers)Check SVG metadata and text alternativeProvide alt text/data table
Convertible to image later✅ anytime— already image
The three-phase workflow
flowchart LR
    accTitle: Three-Phase Documentation Workflow
    accDescr: Markdown holds the structural diagram source. Quantitative charts come from data and plotting code; optional illustrations require a separate factual review.

    p1["📄 Phase 1<br/>Mermaid in Markdown<br/>(structural source)"]
    p2["🐍 Phase 2<br/>Python Generated<br/>(optional — data charts)"]
    p3["🎨 Phase 3<br/>AI Generated Visuals<br/>(optional — polish)"]
    out["📊 Final Deliverable"]

    p1 --> out
    data["Data and plotting code"] --> p2
    p1 -.->|"when needed"| p3
    p2 --> out
    p3 --> out

    classDef required fill:#dbeafe,stroke:#2563eb,stroke-width:2px,color:#1e3a5f
    classDef optional fill:#fef9c3,stroke:#ca8a04,stroke-width:2px,color:#713f12
    classDef output fill:#dcfce7,stroke:#16a34a,stroke-width:2px,color:#14532d

    class p1 required
    class p2,p3 optional
    class out output

Retain Mermaid for structural diagrams and data/code for quantitative charts. An AI illustration is not a quantitative conversion; verify every label and relationship against its sources.

What Mermaid can express

This skill includes 23 diagram-type guides plus a composition guide. Mermaid supports additional types; these references are a curated subset, not an exhaustive version catalogue:

Use caseDiagram typeFile
Experimental workflow / decision logicFlowchartreferences/diagrams/flowchart.md
Service interactions / API calls / messagingSequencereferences/diagrams/sequence.md
Data model / schemaER diagramreferences/diagrams/er.md
State machine / lifecycleStatereferences/diagrams/state.md
Project timeline / roadmapGanttreferences/diagrams/gantt.md
Proportions / compositionPiereferences/diagrams/pie.md
System architecture (zoom levels)C4references/diagrams/c4.md
Concept hierarchy / brainstormMindmapreferences/diagrams/mindmap.md
Chronological events / historyTimelinereferences/diagrams/timeline.md
Class hierarchy / type relationshipsClassreferences/diagrams/class.md
User journey / satisfaction mapUser Journeyreferences/diagrams/user_journey.md
Two-axis comparison / prioritizationQuadrantreferences/diagrams/quadrant.md
Requirements traceabilityRequirementreferences/diagrams/requirement.md
Flow magnitude / resource distributionSankeyreferences/diagrams/sankey.md
Numeric trends / bar + line chartsXY Chartreferences/diagrams/xy_chart.md
Component layout / spatial arrangementBlockreferences/diagrams/block.md
Work item status / task columnsKanbanreferences/diagrams/kanban.md
Cloud infrastructure / service topologyArchitecturereferences/diagrams/architecture.md
Multi-dimensional comparison / skills radarRadarreferences/diagrams/radar.md
Hierarchical proportions / budgetTreemapreferences/diagrams/treemap.md
Binary protocol / data formatPacketreferences/diagrams/packet.md
Git branching / merge strategyGit Graphreferences/diagrams/git_graph.md
Code-style sequence (programming syntax)ZenUMLreferences/diagrams/zenuml.md
Multi-diagram composition patternsComplex Examplesreferences/diagrams/complex_examples.md

💡 Pick the right type, not the easy one. Don't default to flowcharts for everything. A timeline beats a flowchart for chronological events. A sequence beats a flowchart for service interactions. Scan the table and match.


🔧 Core workflow

Step 1: Identify the document type

Check if a template exists before writing from scratch:

Document typeTemplate
Pull request recordtemplates/pull_request.md
Issue / bug / feature requesttemplates/issue.md
Sprint / project boardtemplates/kanban.md
Architecture decision (ADR)templates/decision_record.md
Presentation / briefingtemplates/presentation.md
Research paper / analysistemplates/research_paper.md
Project documentationtemplates/project_documentation.md
How-to / tutorialtemplates/how_to_guide.md
Status reporttemplates/status_report.md
Step 2: Read the style guide

For this Markdown workflow, read references/markdown_style_guide.md. Treat emoji, heading counts, and horizontal rules as house style rather than Markdown syntax requirements.

Key rules to internalize:

  • One H1 per document — the title. Never more.

  • Emoji on H2 headings only — one emoji per H2, none in H3/H4

  • Support factual claims — use verified sources and the requested citation format

  • Bold sparingly — max 2-3 bold terms per paragraph, never full sentences

  • Optional horizontal rules after </details> when they improve separation

  • Tables over prose for comparisons, configurations, structured data

  • Diagrams over walls of text — if it describes flow, structure, or relationships, add Mermaid

Step 3: Pick the diagram type and read its guide

Before creating any Mermaid diagram: read references/mermaid_style_guide.md.

Then open the specific type file (e.g., references/diagrams/flowchart.md) for the exemplar, tips, and copy-paste template.

Add accessibility metadata for types that emit it in the chosen renderer:

accTitle: Short Name 3-8 Words
accDescr: One or two sentences explaining what this diagram shows.
  • Prefer host themes; %%{init} directives are deprecated in favor of YAML configuration

  • Prefer reusable classDef where supported; style syntax is diagram-specific

  • One emoji per node max — at the start of the label

  • Use descriptive IDs with the type's syntax: snake_case works for flowcharts; ER/class/state conventions differ

Step 4: Write the document

Start from the template. Apply the markdown style guide. Place diagrams inline with related text — not in a separate "Figures" section.

Render in the actual destination before delivery. Check its Mermaid version and plugin requirements before choosing newer diagram types; GitHub documents an info diagram for this check. A successful latest-version local preview does not establish GitHub or another host will render it. Where Mermaid is unsupported, supply a rendered SVG/PNG with a text description alongside the retained .md source.

Step 5: Retain source and verification

Keep the .md source, rendered deliverable, and tested renderer version together. Follow the project's version-control workflow; this skill does not itself authorize a commit or publication. For CLI export, embedding, and the tested accessibility matrix, read Current rendering and validation.


⚠️ Common pitfalls

Radar chart syntax (radar-beta)

WRONG (shown as text so it does not break the document renderer):

radar
title Example
x-axis ["A", "B", "C"]
"Series" : [1, 2, 3]

CORRECT:

radar-beta
accTitle: Example Radar Scores
accDescr: Illustrative values of one, two, and three on a shared zero-to-three scale.
title Example
axis a["A"], b["B"], c["C"]
curve series["Series"]{1, 2, 3}
max 3
  • Use radar-beta not radar (the bare keyword doesn't exist)

  • Use axis to define dimensions, not x-axis

  • Use curve to define data series, not quoted labels with colon

  • Mermaid 12.0.0 radar emits accTitle/accDescr; check older hosts and retain a visible description

XY Chart vs Radar confusion
DiagramKeywordAxis syntaxData syntax
XY Chart (bars/lines)xychart-betax-axis ["Label1", "Label2"]bar [10, 20] or line [10, 20]
Radar (spider/web)radar-betaaxis id["Label"]curve id["Label"]{10, 20}
Forgetting accTitle/accDescr on supported types

Only some diagram types support accTitle/accDescr. For those that don't, always place a descriptive italic paragraph directly above the code block:

Radar chart comparing three methods on a stated, common score scale. See the data table for exact values.

radar-beta
...

🔗 Integration with other skills

With scientific-schematics

scientific-schematics generates AI-powered publication-quality images (PNG). Use the Mermaid diagram as the brief for the schematic:

Workflow:
1. Create the concept as Mermaid in .md (this skill — Phase 1)
2. Describe the same concept to scientific-schematics for a polished PNG (Phase 3)
3. Commit both — the .md as source, the PNG as a supplementary figure
With scientific-writing

For a manuscript that uses Mermaid structural figures, this skill handles diagram syntax. Follow the manuscript's required figure formats and scientific visualization conventions.

Workflow:
1. Use scientific-writing to draft the manuscript
2. For every figure that shows a workflow, architecture, or relationship:
   - Replace placeholder with a Mermaid diagram following this skill's guide
3. Use plotting or schematic tools where appropriate to the scientific content
With literature-review

Literature review produces summaries with lots of relationship data. Use this skill to:

  • Create concept maps (Mindmap) of the literature landscape

  • Show publication timelines (Timeline or Gantt)

  • Compare methodologies (Quadrant or Radar)

  • Diagram data flows described in papers (Sequence or Flowchart)

With any skill that produces output documents

Before finalizing a document using this skill, check:

  • [ ] Does the document use a template? If so, did I start from the right one?

  • [ ] Is each diagram in an appropriate format, with working accessibility metadata or a text alternative?

  • [ ] Is configuration supported by the destination and tested in light/dark themes?

  • [ ] Do citations support external claims, and are synthetic values labeled?

  • [ ] One H1, emoji on H2 only?

  • [ ] Does the final Markdown/HTML structure render correctly?


📚 Reference index

Style guides
GuidePathLinesWhat it covers
Markdown Style Guidereferences/markdown_style_guide.mdsee fileHeadings, formatting, citations, tables, Mermaid integration, templates, quality checklist
Mermaid Style Guidereferences/mermaid_style_guide.mdsee fileAccessibility, emoji set, color classes, theme neutrality, type selection, complexity tiers
Current Renderingreferences/current-rendering.mdsee fileMermaid 12 changes, CLI export, browser APIs, tested accessibility matrix, scientific checks
Diagram guides (23 types plus composition)

Each file contains: production-quality exemplar, tips specific to that type, and a copy-paste template.

references/diagrams/ — architecture, block, c4, class, complex_examples, er, flowchart, gantt, git_graph, kanban, mindmap, packet, pie, quadrant, radar, requirement, sankey, sequence, state, timeline, treemap, user_journey, xy_chart, zenuml

Document templates (9 types)

templates/ — decision_record, how_to_guide, issue, kanban, presentation, project_documentation, pull_request, research_paper, status_report

Examples

assets/examples/example-research-report.md — a synthetic quality-control report with explicitly invented data, an auditable count table, a structural flowchart, a bar chart, and a chronology. It makes no experimental or software API claims.


📝 Attribution

All style guides, diagram type guides, and document templates in this skill are ported from the SuperiorByteWorks-LLC/agent-project repository under the Apache-2.0 License.

This skill (as part of scientific-agent-skills) is distributed under the MIT License. The included Apache-2.0 content is compatible for downstream use with attribution retained, as preserved in the file headers throughout this skill.


[^1]: GitHub Blog. (2022). "Include diagrams in your Markdown files with Mermaid." https://github.blog/2022-02-14-include-diagrams-markdown-files-mermaid/

[^2]: Mermaid. "Mermaid Diagramming and Charting Tool." https://mermaid.js.org/

Files

references/markdown_style_guide.md

<!-- Source: https://github.com/SuperiorByteWorks-LLC/agent-project | License: Apache-2.0 | Author: Clayton Young / Superior Byte Works, LLC (Boreal Bytes) -->

Markdown Style Guide

For AI agents: Read this file for all core formatting rules. When creating any markdown document, follow these conventions for consistent, professional output. When a template exists for your document type, start from it — see Templates.

For humans: This guide ensures every markdown document in your project is clean, scannable, well-cited, and renders beautifully on GitHub. Reference it from your AGENTS.md or contributing guide.

Target platform: GitHub-style Markdown; verify the exact host and export pipeline. Reviewed: 2026-10-01 against CommonMark 0.31.2, GFM, and current GitHub formatting documentation.

Heading counts, emoji, separators, and citation placement below are optional house style. They are not Markdown conformance rules and do not override a requested format or an existing project/journal style. Tables and task lists are GFM extensions; footnotes, Mermaid, HTML details, and generated heading anchors depend on the host. Design goal: Clear, professional documents that communicate effectively through consistent structure, meaningful formatting, proper citations, and strategic use of diagrams.


Quick Start for Agents

  1. Identify the document type → Check if a template exists

  2. Structure first → Heading hierarchy, then content

  3. Apply formatting from this guide → Headings, text, lists, tables, images, links

  4. Add citations → Footnote references for all claims and sources

  5. Consider diagrams → Would a Mermaid diagram communicate this better than text?

  6. Add collapsible sections → For supplementary detail, speaker notes, or lengthy context

  7. Verify → Run through the quality checklist


Core Principles

#PrincipleRule
1Answer before they askAnticipate reader questions and address them inline. A great document resolves doubts as they form — the reader finishes with no lingering "but what about...?"
2Scannable firstReaders skim before they read. Use headings, bold, and lists to make the structure visible at a glance.
3Cite everythingEvery claim, statistic, or external reference gets a footnote citation with a full URL. No orphan claims.
4Diagrams over walls of textIf a concept involves flow, relationships, or structure, use a Mermaid diagram alongside the text.
5Generous with informationDon't hide the details — surface them. Use collapsible sections for depth without clutter, but never omit information because "they probably don't need it." If it's relevant, include it.
6Consistent structureSame heading hierarchy, same formatting patterns, same emoji placement across every document.
7One idea per sectionEach heading should cover one topic. If you're covering two ideas, split into two headings.
8Professional but approachableClean formatting, no clutter, no decorative noise — but not stiff or academic. Write like a senior engineer explains to a colleague.

🗂️ Optional repository records

Markdown can preserve portable decision records and report snapshots. The issue, PR, and kanban templates support projects that choose file-based records; they do not change an existing tracker workflow or automatically synchronize with GitHub, GitLab, or another service. Preserve the project's actual source of truth and link to the live record when needed. Writing a file does not create an issue, post a comment, or update a board.

See File conventions for suggested names.


Document Structure

Title and metadata

Every document starts with exactly one H1 title, followed by a brief context line and a separator:

# Document Title Here

_Brief context — project name, date, or purpose in one line_

---
  • One H1 per document — never more

  • Context line in italics — what this document is, when, and for whom

  • Horizontal rule separates metadata from content

Heading hierarchy
LevelSyntaxUseMax per document
H1# TitleDocument title1 (exactly one)
H2## SectionMajor sections4–10
H3### TopicTopics within a section2–5 per H2
H4#### SubtopicSubtopics when needed2–4 per H3
H5+Never use—0

Rules:

  • Never skip levels — don't jump from H2 to H4

  • Emoji in H2 headings — one emoji per H2, at the start: ## 📋 Project Overview

  • No emoji in H3/H4 — keep sub-headings clean

  • Sentence case — ## 📋 Project overview not ## 📋 Project Overview (exception: proper nouns)

  • Descriptive headings — ### Authentication flow not ### Details


Text Formatting

Bold, italic, code
FormatSyntaxWhen to useExample
Bold**text**Key terms, important concepts, emphasisPrimary database handles writes
Italic*text*Definitions, titles, subtle emphasisThe process is called sharding
Code`text`Technical terms, commands, file names, valuesRun npm install to install
Strike~~text~~Deprecated content, correctionsOld approach replaced by v2

Rules:

  • Bold sparingly — if everything is bold, nothing is. Max 2–3 bold terms per paragraph.

  • Don't combine bold and italic (***text***) — pick one

  • Code for anything technical — file names (README.md), commands (git push), config values (true), environment variables (NODE_ENV)

  • Never bold entire sentences — bold the key word(s) within the sentence

Blockquotes

Use blockquotes for definitions, callouts, and important notes:

> **Definition:** A _load balancer_ distributes incoming network traffic
> across multiple servers to ensure no single server bears too much demand.

For warnings and callouts:

> ⚠️ **Warning:** This operation is destructive and cannot be undone.

> 💡 **Tip:** Use `--dry-run` to preview changes before applying.

> 📌 **Note:** This requires admin permissions on the repository.
  • Prefix with emoji + bold label for typed callouts

  • Keep blockquotes to 1–3 lines

  • Don't nest blockquotes (>>)


Lists

When to use each type
List typeSyntaxUse when
Bullet- itemItems have no inherent order
Numbered1. stepSteps must happen in sequence
Checkbox- [ ] itemTracking completion (agendas, checklists)
Formatting rules
  • Indent nested items under the parent item text — commonly 2 spaces after - and 3 after 1. . Marker width matters; a universal 2-space rule breaks ordered lists

  • Parallel structure — every item in a list should have the same grammatical form

  • No period at end unless items are full sentences

  • Keep items concise — if a bullet needs a paragraph, it should be a sub-section instead

  • Max nesting depth: 2 levels — if you need a third level, restructure

✅ Good — parallel structure, concise:

- Configure the database connection
- Run the migration scripts
- Verify the schema changes

❌ Bad — mixed structure, verbose:

- You need to configure the database
- Migration scripts
- After that, you should verify that the schema looks correct

See the [Mermaid Style Guide](mermaid_style_guide.md) for diagram conventions.
  • Meaningful link text — [Mermaid Style Guide] not [click here] or [link]

  • Relative paths for internal links — resolve them from the containing Markdown file, including copied templates

  • Full URLs for external links — always https://

Footnote citations

Cite external factual claims with sources that support them. Use footnotes if the destination supports them; otherwise use inline/reference links or the required citation style. A citation must support the claim, not merely point to a related page. Explicitly label synthetic data and distinguish it from measurements; never invent references.

Markdown was created by John Gruber in 2004 as a lightweight
markup language designed for readability[^1]. GitHub adopted
Mermaid diagram support in February 2022[^2].

[^1]: Gruber, J. (2004). "Markdown." _Daring Fireball_. https://daringfireball.net/projects/markdown/

[^2]: GitHub Blog. (2022). "Include diagrams in your Markdown files with Mermaid." https://github.blog/2022-02-14-include-diagrams-markdown-files-mermaid/

Citation format:

[^N]: Author/Org. (Year). "Title." *Publication*. https://full-url

Rules:

  • Number sequentially — [^1], [^2], [^3] in order of appearance

  • Full URL always included — the reader must be able to reach the source

  • Group all footnotes at the document bottom — under a ## References section or at the very end

  • Every external claim needs one — statistics, quotes, methodologies, tools mentioned

  • Internal project links don't need footnotes — use inline links instead

When the same URL appears multiple times, use reference-style links to keep the text clean:

The [official docs][mermaid-docs] cover all diagram types.
See [Mermaid documentation][mermaid-docs] for the full syntax.

[mermaid-docs]: https://mermaid.js.org/ 'Mermaid Documentation'

Images and Figures

Placement and syntax
![Descriptive alt text for screen readers](images/architecture_overview.png)
_Figure 1: System architecture showing the three-tier deployment model_

Rules:

  • Inline with content — place images where they're relevant, not in a separate "Images" section

  • Descriptive alt text — ![Three-tier architecture diagram] not ![image] or ![screenshot]

  • Italic caption below — *Figure N: What this image shows*

  • Number figures sequentially — Figure 1, Figure 2, etc. if multiple images

  • Relative paths — images/file.png not absolute paths

  • Reasonable file sizes — compress PNGs, use SVG where possible

Image naming convention
{document-slug}_{description}.{ext}

Examples:
  auth_flow_overview.png
  deployment_architecture.svg
  api_response_example.png
When NOT to use an image

If the content could be expressed as a Mermaid diagram, prefer that over a static image:

ScenarioUse
Architecture diagramMermaid flowchart or architecture-beta
Sequence/interactionMermaid sequenceDiagram
Data modelMermaid erDiagram
TimelineMermaid timeline or gantt
Screenshot of UIImage (Mermaid can't do this)
Photo / real-world imageImage
Complex data visualizationScientific plotting with source data/code

See the Mermaid Style Guide for diagram type selection and styling.


Tables

When to use tables
  • Structured comparisons — features, options, tradeoffs

  • Reference data — configuration values, API parameters, status codes

  • Schedules and matrices — timelines, responsibility assignments

When NOT to use tables
  • Narrative content — use paragraphs instead

  • Simple lists — use bullet points

  • More than 5 columns — becomes unreadable on mobile; restructure

Formatting
| Feature | Free Tier | Pro Tier | Enterprise |
| ------- | --------- | -------- | ---------- |
| Users   | 5         | 50       | Unlimited  |
| Storage | 1 GB      | 100 GB   | Custom     |
| Support | Community | Email    | Dedicated  |

Rules:

  • Header row always — no headerless tables

  • Left-align text columns — |---| (default)

  • Right-align number columns — |---:| when appropriate

  • Concise cell content — 1–5 words per cell. If you need more, it's not a table problem

  • Bold key column — the first column or the column the reader scans first

  • Consistent formatting within columns — don't mix sentences and fragments


Code Blocks

Inline code

Use backticks for technical terms within prose:

Run `git status` to check for uncommitted changes.
The `NODE_ENV` variable controls the runtime environment.
Fenced code blocks

Always specify the language for syntax highlighting:

```python
def calculate_average(values: list[float]) -> float:
    """Return the arithmetic mean of a list of values."""
    return sum(values) / len(values)
```

Rules:

  • Always include language identifier — ```python, ```bash, ```json, etc.

  • Use ```text for plain output — not ``` with no language

  • Keep blocks focused — show the relevant snippet, not the entire file

  • Add a comment if context needed — # Configure the database connection at the top of the block


Collapsible Sections

Use HTML <details> for supplementary content that shouldn't clutter the main flow — speaker notes, implementation details, verbose logs, or optional deep-dives.

<details>
<summary><strong>💬 Speaker Notes</strong></summary>

- Key talking point one
- Transition to next topic
- **Bold** emphasis works inside details
- [Links](https://example.com) work too

</details>

---

Rules:

  • Collapsed by default — the <details> tag collapses automatically

  • Descriptive summary — <strong>💬 Speaker Notes</strong> or <strong>📋 Implementation Details</strong>

  • Blank line after <summary> tag — required for markdown to render inside the block

  • Optional house style: follow with --- if the extra separator improves clarity

  • GitHub supports Markdown inside details with blank lines around the Markdown content. Other renderers may sanitize HTML or leave it unparsed; preview the exported result

Common collapsible patterns
Summary labelUse for
💬 Speaker NotesPresentation talking points, timing, transitions
📋 DetailsExtended explanation, verbose context
🔧 ImplementationTechnical details, code samples, config
📊 Raw DataFull output, logs, data tables
💡 BackgroundContext that helps but isn't essential

Horizontal Rules

Use --- (three hyphens) for visual separation:

---

When to use:

  • After every </details> block — mandatory, creates clear separation

  • After title/metadata — separates document header from content

  • Between major sections — when an H2 heading alone doesn't create enough visual break

  • Before footnotes/references — separates content from citation list

When NOT to use:

  • Between every paragraph (too busy)

  • Between H3 sub-sections within the same H2 (use whitespace instead)


Approved Emoji Set

One emoji per H2 heading, at the start. Use sparingly in body text for callouts and emphasis only.

Section headings
EmojiUse for
📋Overview, summary, agenda, checklist
🎯Goals, objectives, outcomes, targets
📚Content, documentation, main body
🔗Resources, references, links
📍Agenda, navigation, current position
🏠Housekeeping, logistics, announcements
✍️Tasks, assignments, action items
Status and outcomes
EmojiMeaning
✅Success, complete, correct, approved
❌Failure, incorrect, avoid, rejected
⚠️Warning, caution, important notice
💡Tip, insight, idea, best practice
📌Important, key point, remember
🚫Prohibited, do not, blocked
Technical and process
EmojiMeaning
⚙️Configuration, settings, process
🔧Tools, utilities, setup
🔍Analysis, investigation, review
📊Data, metrics, analytics
📈Growth, trends, improvement
🔄Cycle, refresh, iteration
⚡Performance, speed, quick action
🔐Security, authentication, privacy
🌐Web, API, network, global
💾Storage, database, persistence
📦Package, artifact, deployment
People and collaboration
EmojiMeaning
👤User, person, individual
👥Team, group, collaboration
💬Discussion, comments, speaker notes
🎓Learning, education, knowledge
🤔Question, consideration, reflection
Emoji rules
  1. One per H2 heading at the start — ## 📋 Overview

  2. None in H3/H4 — keep sub-headings clean

  3. Sparingly in body text — for callouts (> ⚠️ **Warning:**) and key markers only

  4. Never in: titles (H1), code blocks, link text, table data cells

  5. No decorative emoji — 🎉 💯 🔥 🎊 💥 ✨ add noise, not meaning

  6. Consistency — same emoji = same meaning across all documents in the project


Mermaid Diagram Integration

Use Mermaid for structural relationships when it improves understanding. Use data-driven plotting for quantitative figures that need exact geometry, uncertainty, or a journal-specific format.

When to add a diagram

If a diagram helps the reader, use the table to select a type and verify host support:

  1. Read the Mermaid Style Guide first — emoji, color palette, accessibility, complexity management

  2. Then open the specific type file — exemplar, tips, template, complex example

Your content describes...Add a...Type file
Steps in a process, workflow, decision logicFlowchartflowchart.md
Who talks to whom and when (API calls, messages)Sequence diagramsequence.md
Class hierarchy, type relationships, interfacesClass diagramclass.md
Status transitions, entity lifecycle, state machineState diagramstate.md
Database schema, data model, entity relationshipsER diagramer.md
Project timeline, roadmap, task dependenciesGantt chartgantt.md
Parts of a whole, proportions, distributionPie chartpie.md
Git branching strategy, merge/release flowGit Graphgit_graph.md
Concept hierarchy, brainstorm, topic mapMindmapmindmap.md
Chronological events, milestones, historyTimelinetimeline.md
User experience, satisfaction scores, journeyUser Journeyuser_journey.md
Two-axis comparison, prioritization matrixQuadrant chartquadrant.md
Requirements traceability, compliance mappingRequirement diagramrequirement.md
System architecture at varying zoom levelsC4 diagramc4.md
Flow magnitude, resource distribution, budgetsSankey diagramsankey.md
Numeric trends, bar charts, line chartsXY Chartxy_chart.md
Component layout, spatial arrangement, layersBlock diagramblock.md
Work item tracking, status board, task columnsKanban boardkanban.md
Binary protocol layout, data packet formatPacket diagrampacket.md
Cloud infrastructure, service topology, networkingArchitecture diagramarchitecture.md
Multi-dimensional comparison, skills, radar analysisRadar chartradar.md
Hierarchical proportions, budget breakdownTreemaptreemap.md

💡 Pick the right type, not the easy type. Don't default to flowcharts for everything — a timeline is better than a flowchart for chronological events, a sequence diagram is better for service interactions, an ER diagram is better for data models. Scan the table above and match your content to the most specific type. A short paragraph or data table can be clearer for a small amount of information.

How to integrate

Place the diagram inline with the related text, not in a separate section:

### Authentication Flow

The login process validates credentials, checks MFA status,
and issues session tokens. Failed attempts are logged for
security monitoring.

```mermaid
sequenceDiagram
accTitle: Login Authentication Flow
accDescr: User login sequence through API and auth service

    participant U as 👤 User
    participant A as 🌐 API
    participant S as 🔐 Auth Service

    U->>A: POST /login
    A->>S: Validate credentials
    S-->>A: ✅ Token issued
    A-->>U: 200 OK + session

```

The token expires after 24 hours. See [Authentication flow](#authentication-flow)
for refresh token details.

Always follow the Mermaid Style Guide for diagram styling — emoji, color classes, accessibility (accTitle/accDescr), and type-specific conventions.


Whitespace and Spacing

  • Blank line between paragraphs — always

  • Blank line before and after headings — always

  • Blank line before and after code blocks — always

  • Blank line before and after blockquotes — always

  • No blank line between list items — keep lists tight

  • No trailing whitespace — clean line endings

  • One blank line at end of file — standard convention

  • No more than one consecutive blank line — two blank lines = too much space


Quality Checklist

Structure
  • [ ] Exactly one H1 title

  • [ ] Heading hierarchy is correct (H1 → H2 → H3 → H4, no skips)

  • [ ] Each H2 has exactly one emoji at the start

  • [ ] H3 and H4 have no emoji

  • [ ] Horizontal rules after title metadata and after every </details> block

Content
  • [ ] Every external claim has a footnote citation

  • [ ] All footnotes have full URLs

  • [ ] All links tested and working

  • [ ] Meaningful link text (no "click here")

  • [ ] Bold used for key terms, not entire sentences

  • [ ] Code formatting for all technical terms

Visual elements
  • [ ] Images have descriptive alt text

  • [ ] Images have italic figure captions

  • [ ] Images placed inline with related content (not in separate section)

  • [ ] Tables have header rows and consistent formatting

  • [ ] Mermaid diagrams considered where applicable (with accTitle/accDescr)

Collapsible sections
  • [ ] <details> blocks have descriptive <summary> labels

  • [ ] Blank line after <summary> tag (for markdown rendering)

  • [ ] Horizontal rule --- after every </details> block

  • [ ] Content inside collapses renders correctly

Polish
  • [ ] No spelling or grammar errors

  • [ ] Consistent whitespace (no trailing spaces, no double blanks)

  • [ ] Parallel grammatical structure in lists

  • [ ] Renders correctly in GitHub light and dark mode


Templates

Templates provide pre-built structures for common document types. Copy the template, fill in your content, and follow this style guide for formatting. Every template enforces the principles above — citations, diagrams, collapsible depth, and self-answering structure.

Document typeTemplateBest for
Presentation / briefingpresentation.mdSlide-deck-style documents with speaker notes, structured sections, and visual flow
Research paper / analysisresearch_paper.mdData-driven analysis, literature reviews, methodology + findings with heavy citations
Project documentationproject_documentation.mdSoftware/product docs — architecture, getting started, API reference, contribution guide
Decision record (ADR/RFC)decision_record.mdRecording why a decision was made — context, options evaluated, outcome, consequences
How-to / tutorial guidehow_to_guide.mdStep-by-step instructions with prerequisites, verification steps, and troubleshooting
Status report / executive briefstatus_report.mdProgress updates, risk summaries, decisions needed — for leadership and stakeholders
Pull request recordpull_request.mdPR documentation with change inventory, testing evidence, rollback plan, and review notes
Issue recordissue.mdBug reports (reproduction steps, root cause) and feature requests (acceptance criteria, user stories)
Kanban boardkanban.mdSprint/release/project work tracking with visual board, WIP limits, metrics, and blocked items
File conventions for tracked documents

Some templates produce documents that accumulate over time. Use these directory conventions:

Document typeDirectoryNaming patternExample
Pull requestsdocs/project/pr/pr-NNNNNNNN-short-description.mddocs/project/pr/pr-00000123-fix-auth-timeout.md
Issuesdocs/project/issues/issue-NNNNNNNN-short-description.mddocs/project/issues/issue-00000456-add-export-filter.md
Kanban boardsdocs/project/kanban/{scope}-{identifier}-short-description.mddocs/project/kanban/sprint-2026-w07-agentic-template-modernization.md
Decision recordsdocs/decisions/NNN-{slug}.mddocs/decisions/001-use-postgresql.md
Status reportsdocs/status/status-{date}.mddocs/status/status-2026-02-14.md
Choosing a template
  • Presenting to people? → Presentation

  • Publishing analysis or research? → Research paper

  • Documenting a codebase or product? → Project documentation

  • Recording why you chose X over Y? → Decision record

  • Teaching someone how to do something? → How-to guide

  • Updating leadership on progress? → Status report

  • Documenting a PR for posterity? → Pull request record

  • Tracking a bug or requesting a feature? → Issue record

  • Managing work items for a sprint or project? → Kanban board

  • None of these fit? → Start from this style guide's rules directly — no template required


Common Mistakes

❌ Multiple emoji per heading
## 📚📊📈 Content Topics ← Too many

✅ Fix: One emoji per H2

## 📚 Content topics
❌ Missing citations
The renderer supports every diagram type. ← Which renderer/version?

✅ Fix: Add footnote

GitHub documents Mermaid code blocks and an `info` diagram for checking its Mermaid version[^1].

[^1]: GitHub Docs. "Creating diagrams." https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/creating-diagrams
❌ Wall of text without structure
The system handles authentication by first checking the JWT token
validity, then verifying the user exists in the database, then
checking their permissions against the requested resource...

✅ Fix: Use a list, heading, or diagram

### Authentication flow

1. Validate JWT token signature and expiration
2. Verify user exists in the database
3. Check user permissions against the requested resource
❌ Images in a separate section
## Content

[paragraphs of text]

## Screenshots

[all images grouped here] ← Disconnected from context

✅ Fix: Place images inline where relevant

❌ No horizontal rule after collapsible sections
</details>
### Next Topic  ← Runs together visually

✅ Fix: Always add --- after </details>

</details>

---

### Next topic ← Clear separation

Resources

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Declares allowed tools, Needs Node.js

The author says: Markdown authoring needs no runtime. Local rendering targets Mermaid and Mermaid CLI 12.0.0, Node.js 22.13+ and a supported Chromium browser. Package or CDN installation needs network access; no credentials required.

Security

What our scanner noticed· Low risk

  • Instructions that use the network

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by
Clayton Young / Superior Byte Works, LLC (@borealBytes)
Publisher
K-Dense (Community)
Distributed by
K-Dense
Source
K-Dense-AI/scientific-agent-skills · skills/markdown-mermaid-writing · commit a25cf9e
Licence
Apache License 2.0 (redistribution allowed)
  • LICENSE (written by AutoGineer)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by Clayton Young / Superior Byte Works, LLC (@borealBytes) or K-Dense.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download markdown-mermaid-writing.zip