RA

radar-doc-review

Manual documentation reviewer ensuring style guide and format compliance.

Install

mkdir -p .claude/skills/radar-doc-review && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/10423" && unzip -o skill.zip -d .claude/skills/radar-doc-review && rm skill.zip

Installs to .claude/skills/radar-doc-review

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Manually invoked documentation reviewer. Do NOT load this skill automatically. Only load when the user explicitly runs /radar-doc-review. Reviews documentation files for correct format, style guide compliance, and content quality.
230 chars✓ has a “when” trigger
Intermediate

Key capabilities

  • Review documentation structure
  • Check style guide compliance
  • Validate link integrity
  • Score documentation quality

How it works

It reviews files against a style guide and example document, scoring them on content, format, and SEO.

Inputs & outputs

You give it
Markdown documentation file
You get back
Structured quality report

When to use radar-doc-review

  • Reviewing new documentation files
  • Ensuring style guide compliance
  • Auditing doc quality

About this skill

Documentation Review Skill

Purpose

Review documentation files against the project's style guide, doc-type format requirements, and the established example doc.

Invocation

This skill is manually invoked only via /radar-doc-review. Do not auto-load this skill based on context.

Supported file types: This skill reviews .mdx and .md files only. If the user provides a different file type, notify them that the skill only supports Markdown-based documentation files.


Step-by-Step Review Process

Step 1 — Read the Reference Files

Before doing anything else, read these files in full:

  1. Style guide: /docs/style-guide for .md files in all directories.
  2. Example doc: /content/hcp-docs/content/docs/vault-radar/get-started/add-data-sources/slack.mdx
    • This is a generally well-formatted doc. Note its frontmatter fields, heading hierarchy, tone, code block usage, admonition style, link formatting, and overall structure.

Error handling: If either file cannot be found, stop the review and report the missing file path to the user.

Do not proceed until you have read both files.

Step 2 — Identify the Doc Type

Determine which of the following doc types the file belongs to. These align with the content types defined in /docs/content-guide/content-types.md.

Doc TypePrimary GoalAudience Assumption
TutorialTeaches users about product features through a hands-on experience with an end-to-end workflowNo prior knowledge assumed
How-toWalks users through completing a specific taskSome prior knowledge assumed
ConceptProvides context and background to help readers understand a product, feature, or topicCurious, not necessarily doing a task
ReferenceTechnical details like API endpoints, CLI commands, and configuration optionsUser knows what they're looking for
TroubleshootingHelps users resolve common issuesUser is encountering a problem
Landing pageProvides an overview of a section and links to child pagesNavigating or exploring
Release notesCommunicates new features, bug fixes, and changesTracking product changes

If you cannot determine the doc type from the file content or frontmatter, ask the user before continuing.

Step 3 — Check Format for the Doc Type

Use the checklist for the identified doc type below. Use the slack.mdx example as a concrete model for what correct formatting looks like.

Tutorial

  • Frontmatter includes: page_title, description, and any required metadata per the style guide
  • Starts with a brief intro explaining what the user will accomplish and any prerequisites
  • Numbered steps throughout — no bullet points for procedural content
  • Step numbers render in the correct order. If steps under a single section reset to 1, identify the offending component that is not indented properly.
  • Each step has exactly one action; compound steps are split unless it is navigating a menu with multiple levels.
  • Ends with a "Next steps" or "What's next" section

How-to

  • Frontmatter includes: page_title, description, and any required metadata
  • Title follows "How to [verb] [noun]" pattern or equivalent action-oriented pattern
  • Brief intro states the goal and any prerequisites
  • Numbered steps for any procedural content
  • Does not over-explain concepts — stays task-focused
  • Optional but encouraged: troubleshooting section at the end

Concept

  • Frontmatter includes: page_title, description, and any required metadata
  • Opens with a clear definition or "what is X" statement
  • Uses prose paragraphs, not numbered steps
  • Explains why something exists or works the way it does
  • Links to related tutorials or how-to guides for hands-on follow-up
  • Does not include step-by-step instructions (those belong in tutorials/how-tos)

Reference

  • Frontmatter includes: page_title, description, and any required metadata
  • Highly structured: uses consistent heading hierarchy and tables where applicable
  • Each item/entry is complete and self-contained
  • No prose narrative — scannable by design
  • Parameters, flags, or fields include: name, type, required/optional, description, and default value (where applicable)
  • No tutorial-style steps or conceptual explanations inline

Troubleshooting

  • Frontmatter includes: page_title, description, and any required metadata
  • Title clearly indicates this is troubleshooting content (e.g., "Troubleshoot [feature]")
  • Organized by symptom or error message — users should find their problem quickly
  • Each issue follows a consistent pattern: symptom/error → cause → resolution
  • Resolution steps are numbered and actionable
  • Does not mix conceptual explanations into resolution steps
  • Links to related how-to or reference docs where relevant

Landing page

  • Frontmatter includes: page_title, description, and any required metadata
  • Provides a brief overview of the section or topic area
  • Links to all relevant child pages with short descriptions
  • Does not contain detailed procedural or conceptual content
  • Organized logically (e.g., by workflow order, complexity, or category)
  • Uses consistent formatting for link lists or card layouts

Release notes

  • Frontmatter includes: page_title, description, and any required metadata
  • Clearly states the version or date of the release
  • Organizes changes by category (e.g., Features, Improvements, Bug fixes, Breaking changes, Deprecations)
  • Each entry is concise — one to two sentences per item
  • Breaking changes and deprecations are prominently highlighted (e.g., using <Warning> or <Note>)
  • Links to relevant docs for new features or changed behavior
  • Does not include tutorial-style instructions

All doc types (universal checks)

  • Heading hierarchy is correct: H1 → H2 → H3, no skipped levels
  • Code blocks have a language identifier (```bash, ```hcl, ```json, etc.)
  • No orphaned H1s — only one H1 per file (the page title)
  • Images (if any) have descriptive alt text
  • Links use the correct format per the style guide (relative vs. absolute)
  • Admonitions/callouts use the correct syntax and are used appropriately (note, warning, tip)
  • No broken links or references to non-existent sections
  • Optimized for SEO including action oriented headings, titles, and descriptions
  • Line break after each section, paragraph, unordered list, ordered list, code blocks, headings, at the end of the document, and other elements for readability

Step 4 — Check Style Guide Compliance

Using the style guide in /docs/style-guide/, check for compliance with the following rules. Flag every violation found.

Voice, tone, and point of view

  • Use second person ("you") — never "we", "our", or "let's"
  • Use active voice — avoid passive constructions (e.g., "the secret is stored" → "Vault stores the secret")
  • Use present tense — avoid future tense ("will"). Write "the command returns" not "the command will return"
  • Use imperative mood for instructions — "Run the command" not "You should run the command"
  • Do not use "please" in instructions
  • Do not use "simple", "easy", "just", or other minimizing language

Terminology and product names

  • Flag any terms the style guide marks as preferred, avoided, or with specific casing
  • HashiCorp product names must be capitalized correctly (e.g., "Vault", "HCP Vault Radar", "Terraform")
  • For HCP product names, use the full name on first reference, then the shortname after
    • "HCP Vault Radar" then "Vault Radar"
    • "HCP Vault Dedicated" then "HCP Vault"
  • Spell out acronyms on first use, then abbreviate (e.g., "Key-Value (KV) secrets engine" then "KV" thereafter)
  • For non-HashiCorp products, use the correct capitalization and spelling per the vendor's guidelines (e.g., "Slack", not "slack" or "SLACK")
  • Do not use Latin abbreviations: write "for example" not "e.g.", "that is" not "i.e.", avoid "etc."

Formatting

  • UI elements: Bold for UI labels (e.g., Save, Settings)
  • Code elements: Use code formatting for commands, values, file paths, API endpoints, and configuration keys
  • Placeholders: Use ALL_CAPS for user-supplied values (e.g. YOUR_TOKEN)
  • Bold and italics: Do not overuse — bold for emphasis or UI, italics sparingly

Headings

  • Use sentence case for all headings
  • Do not start headings with gerunds (-ing words)
  • Do not start headings with articles (a, an, the)
  • Keep headings under 12 words
  • Headings must be action-oriented for procedural content

Capitalization and punctuation

  • Use Oxford commas (serial commas)
  • Use sentence case for headings, titles, and descriptions
  • Spell out numbers under 10; use numerals for 10 and above

Sentence and paragraph structure

  • Flag sentences over ~30 words as candidates for splitting
  • Use shorter, more common words where possible (e.g., "use" not "utilize", "start" not "initiate")
  • Avoid jargon without explanation

Links

  • Use descriptive link text — never "click here" or "this page"
  • Use relative links for internal cross-references where the style guide specifies
  • Verify link text accurately describes the target

Alerts and admonitions

  • Use the correct component for the context:
    • <Note>: Supplementary information that is useful but not critical
    • <Tip>: Helpful suggestion or best practice
    • <Warning>: Information that could cause data loss, security issues, or breaking changes
    • <Highlight>: Important callout for limitations or key details
  • Do not overuse admonitions — if everything is highlighted, nothing stands out

Lists

  • Use numbered lists for sequential/procedural steps
  • Use **bull

Content truncated.

When not to use it

  • Auto-loading in context

Limitations

  • Supports only .md and .mdx files
  • Requires manual invocation

How it compares

This provides a formal, score-based review process compared to informal peer review.

Compared to similar skills

radar-doc-review side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
radar-doc-review (this skill)05moReviewIntermediate
ml-paper-writing486moReviewAdvanced
docs-review107moNo flagsBeginner
claude-md-improver216moReviewBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

ml-paper-writing

davila7

Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.

4897

docs-review

metabase

Review documentation changes for compliance with the Metabase writing style guide. Use when reviewing pull requests, files, or diffs containing documentation markdown files.

1085

claude-md-improver

anthropics

Audit and improve CLAUDE.md files in repositories. Use when user asks to check, audit, update, improve, or fix CLAUDE.md files. Scans for all CLAUDE.md files, evaluates quality against templates, outputs quality report, then makes targeted updates. Also use when the user mentions "CLAUDE.md maintenance" or "project memory optimization".

2167

write-docs

tldraw

Writing SDK documentation for tldraw. Use when creating new documentation articles, updating existing docs, or when documentation writing guidance is needed. Applies to docs in apps/docs/content/.

665

update-docs

vercel

This skill should be used when the user asks to "update documentation for my changes", "check docs for this PR", "what docs need updating", "sync docs with code", "scaffold docs for this feature", "document this feature", "review docs completeness", "add docs for this change", "what documentation is affected", "docs impact", or mentions "docs/", "docs/01-app", "docs/02-pages", "MDX", "documentation update", "API reference", ".mdx files". Provides guided workflow for updating Next.js documentation based on code changes.

2543

wiki-architect

microsoft

Analyzes code repositories and generates hierarchical documentation structures with onboarding guides. Use when the user wants to create a wiki, generate documentation, map a codebase structure, or understand a project's architecture at a high level.

1144

Search skills

Search the agent skills registry