agentleFS
jggomezjggomez/expert-ai-developer-skills/skills/documentation-expert/SKILL.md

documentation-expert

Skill30 starsChanged 4 days agoApache-2.0View on GitHub

Author and validate Zero-Fluff technical documentation, root AGENTS.md, Google DESIGN.md alignment, Diátaxis guides, and Mermaid diagrams with strict conciseness caps and relative link verification.

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 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.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use concise Mermaid blocks (flowchart TD, flowchart LR, sequenceDiagram, stateDiagram-v2, erDiagram) instead of prose walls.

    line 32
  2. Distinctive:

    Use portable repository-relative links (never machine-local paths) and run the validator script:

    line 35
  3. Very specific:

    Claude Code: Run validate_docs.py via Bash after authoring or editing markdown docs.

    line 48
  4. Very specific:

    Antigravity: Verify Mermaid quoting, line caps, and relative links via validate_docs.py.

    line 49

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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

Documentation Expert Skill

Overview

This skill guides the authoring, structuring, and automated validation of Zero-Fluff technical documentation, root AGENTS.md guides, official Google DESIGN.md references, and Mermaid.js architectural diagrams. Applying the Diátaxis framework (Tutorials, How-To Guides, Reference, Explanation), it enforces strict line budgets, portable relative links, and zero-boilerplate clarity.

When to Use

Trigger Scenarios

  • Authoring or auditing READMEs, runbooks, API references, and root AGENTS.md (<= 90 lines).
  • Ensuring documentation aligns with official Google 2-layer DESIGN.md (<= 110 lines) and docs/<feature-name>/ line caps.
  • Modeling architectures, sequence flows, or schemas with syntax-safe Mermaid.js diagrams.
  • Validating markdown heading hierarchy, relative links, and path portability.

When NOT to Use

  • Authoring MADR architectural decisions (adr-NNN-*.md): Route to senior-architect-engineering.
  • Authoring UI token systems (DESIGN.md): Route to ui-ux-designer.
  • Automated codebase cartography (project-context.md): Route to repo-research.

Process

Phase 1: Diátaxis Pillar & Zero-Fluff Budget Alignment

  1. Select one Diátaxis mode per section — never mix tutorials, how-to steps, reference tables, and explanations together.
  2. Enforce strict Zero-Fluff conciseness caps across project and feature artifacts:
    • Root AGENTS.md <= 90 lines (verified build/lint/test commands, conventions, links to DESIGN.md & context/memory/MEMORY.md).
    • Root Google DESIGN.md <= 110 lines (YAML tokens + 8 sections) and feature delta docs/<feature-name>/design.md <= 35 lines.
    • Feature specs in docs/<feature-name>/: prd.md <= 70 lines, adr-NNN-*.md <= 55 lines, architecture.md <= 70 lines, test-matrix.md <= 80 lines, verification reports <= 50 lines.

Phase 2: Syntax-Safe Mermaid.js Diagramming

Use concise Mermaid blocks (flowchart TD, flowchart LR, sequenceDiagram, stateDiagram-v2, erDiagram) instead of prose walls. Always quote labels containing spaces, slashes, brackets, or parentheses (e.g., api["Auth API (v2)"]).

Phase 3: Link Portability & Automated Validation

Use portable repository-relative links (never machine-local paths) and run the validator script:

python3 ./skills/documentation-expert/scripts/validate_docs.py

Usage

Commands & Automation Scripts

# Audit markdown heading hierarchy, relative links, and local path leaks
python3 ./skills/documentation-expert/scripts/validate_docs.py

Host Execution Instructions

  • Claude Code: Run validate_docs.py via Bash after authoring or editing markdown docs.
  • Antigravity: Verify Mermaid quoting, line caps, and relative links via validate_docs.py.

Red Flags

  • Bloated prose walls exceeding artifact line budgets (AGENTS.md > 90, DESIGN.md > 110, reports > 50 lines).
  • Blending How-To recipes with deep theoretical Explanations in the same section.
  • Leaking local machine paths or leaving broken relative .md links.
  • Unquoted parentheses or brackets in Mermaid node labels causing render failures.

Verification

  • Document matches its target Diátaxis pillar and stays within its Zero-Fluff line cap.
  • Root AGENTS.md and DESIGN.md cross-links resolve cleanly when present.
  • Validator script exits with code 0:
    python3 ./skills/documentation-expert/scripts/validate_docs.py
  • Zero broken relative links, skipped heading levels, or local path leaks.

References

  • Diátaxis Documentation Framework Reference
  • Mermaid.js Graphic Reference Guide

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.