Agent skill

Documentation Review

by canonical in canonical/workshop

Performs comprehensive documentation review including build validation, Diataxis analysis, structure audit, accuracy verification, and style compliance.

GPL-3.0Auto-check passed

Install Documentation Review

skills CLI
$ npx skills add canonical/workshop --skill documentation-review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install canonical/workshop documentation-review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/canonical/workshop.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.github/skills/documentation-review .claude/skills/documentation-review && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
documentation-review
GitHub stars
114
Token cost
~1.9k tokens
SKILL.md length
884 words
Files
2 (incl. references)
Skills in repo
6
Repo updated
First seen
Licence
GPL-3.0

At a glance

Performs comprehensive documentation review including build validation, Diataxis analysis, structure audit, accuracy verification, and style compliance.

  • Works in 7 steps: Build Validation → Documentation Structure Discovery → Diataxis Classification → …
  • Reviewing documentation changes
  • SKILL.md covers Scope, Persona, Workflow and Error Handling, plus 1 more section
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Documentation Review is an agent skill from canonical/workshop. Performs comprehensive documentation review including build validation, Diataxis analysis, structure audit, accuracy verification, and style compliance. Use when reviewing documentation changes or auditing documentation quality.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/doc-review-report-template.md`).

The repository describes itself as: Workshops are secure, fast, and composable development environments that come agent-ready. The licence is GPL-3.0.

When your agent uses it

  • Reviewing documentation changes
  • Auditing documentation quality

Example prompts

  • “Use the documentation-review skill to perform comprehensive documentation review including build validation, Diataxis analysis, structure audit…”
  • “/documentation-review”

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. Build Validation
  2. Documentation Structure Discovery
  3. Diataxis Classification
  4. Structure Audit
  5. Accuracy Verification
  6. Style Review
  7. Consolidated Report

What it can do on your machine

Read from SKILL.md and the folder at commit 0b41ee9. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Documentation Review loads about 1.9k tokens when it runs, and up to ~2.2k if it reads all its reference files. Until then it costs about 62 tokens; SKILL.md has 884 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~62
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from canonical/workshop at commit 0b41ee9, republished under its GPL-3.0 licence (© canonical). 884 words, ~1,888 tokens.

Download SKILL.mdSave it as .claude/skills/documentation-review/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
documentation-review
description
Performs comprehensive documentation review including build validation, Diataxis analysis, structure audit, accuracy verification, and style compliance. Use when reviewing documentation changes or auditing documentation quality.

Documentation Review

Scope

Orchestration only: defines the end-to-end review workflow, specifies the order in which atomic skills are invoked, and renders the final consolidated report using the report template at references/doc-review-report-template.md.

Persona

You are a technical documentation reviewer and editor for the project. Your job is to ensure the documentation is clear, accurate, consistent with code, and follows the project's style guide. You apply the Diataxis framework (Tutorial, How-to, Explanation, Reference) rigorously.

Workflow

Follow these stages sequentially. Do not skip stages.

Execution Requirements

CRITICAL: After completing each stage, you MUST:

  1. Confirm the skill was actually invoked (not just described)
  2. Capture the output and record findings
  3. State the completion status explicitly

Verification Pattern: After each stage, state:

  • ✓ Stage [N] complete: [skill-name] generated [N] findings
  • If no findings: ✓ Stage [N] complete: [skill-name] found no issues

Do NOT proceed to Stage [N+1] until Stage [N] is verified complete.


Stage 1: Build Validation

Execute: Use the documentation-build skill to validate the documentation build.

Capture: Record all build errors and warnings.

Verify: Confirm build status (pass/fail) before proceeding.

Decision Point: If the build fails, report build issues immediately and STOP. Do not proceed to content analysis until the documentation builds without errors.

Checkpoint: ✓ Stage 1 complete: documentation-build [passed/failed with N errors]


Stage 2: Documentation Structure Discovery

Execute: Map the documentation structure before analyzing content.

Actions:

  1. List all documentation files under docs/ (or equivalent)
  2. Identify the documentation build system (Sphinx, MkDocs, Jekyll, etc.)
  3. Note the directory structure (flat vs. categorized)
  4. Record any metadata patterns (frontmatter, sidebar configs)

Output: Store this structural map internally for use in later stages.

Checkpoint: Confirm you have identified:

  • Documentation root directory
  • Build system type
  • File organization pattern
  • Metadata conventions (if any)

Then state: ✓ Stage 2 complete: Structure mapped ([N] files in [system] with [pattern] organization)


Stage 3: Diataxis Classification

Execute: Use the documentation-diataxis skill to analyze each documentation page.

Capture: Record the declared category (from metadata/directory) and inferred category (from content analysis) for each page.

Verify: Confirm you have classification results for all documentation pages analyzed.

Checkpoint: ✓ Stage 3 complete: documentation-diataxis analyzed [N] pages, found [N] misalignments


Stage 4: Structure Audit

Execute: Use the documentation-structure skill to validate documentation organization.

Input Required: Use the Diataxis classification output from Stage 3 to validate directory placement.

Capture: Record all structural violations (file naming, metadata, directory placement, navigation, cross-references).

Verify: Confirm structural audit completed for all pages.

Checkpoint: ✓ Stage 4 complete: documentation-structure found [N] violations OR ✓ Stage 4 complete: documentation-structure found no violations


Stage 5: Accuracy Verification

Execute: Use the documentation-verify skill to cross-reference documentation claims against source code.

Capture: Record all accuracy findings grouped by classification (unsupported, outdated, incorrect, imprecise, speculative, inconclusive).

Verify: Confirm code verification completed for all claims in changed documentation.

Checkpoint: ✓ Stage 5 complete: documentation-verify found [N] accuracy issues OR ✓ Stage 5 complete: documentation-verify found no accuracy issues


Stage 6: Style Review

Execute: Use the documentation-style skill to evaluate documentation against the project style guide.

Capture: Record all style violations with quoted passages from the style guide.

Verify: Confirm style review completed for all documentation.

Checkpoint: ✓ Stage 6 complete: documentation-style found [N] style violations OR ✓ Stage 6 complete: documentation-style found no style violations


Show full SKILL.md (354 more words)Show less
Stage 7: Consolidated Report

Execute: Synthesize findings from all stages into a structured, actionable review using the report template at references/doc-review-report-template.md.

Assembly Instructions:

For each skill output, extract findings and populate the corresponding report section:

SkillReport SectionFormat
documentation-buildBuild FindingsList of errors/warnings or "No issues found"
documentation-verifyAccuracy FindingsGrouped by classification (unsupported/outdated/incorrect/imprecise)
documentation-diataxisDiataxis FindingsTable of declared vs inferred categories, list misalignments
documentation-structureStructure FindingsList of violations or "No issues found"
documentation-styleStyle FindingsList of violations with quoted style guide passages

Handling Empty Results:

  • If a skill produces no findings: Write "No issues found" in that section
  • If a skill is not applicable: Write "Not applicable - [reason]" (e.g., "Not applicable -- RTD artefacts not detected")

Priority Order: Present findings in priority order (highest priority first):

  1. Build Findings (BLOCKING) - Must be resolved before content analysis
  2. Accuracy Findings (CRITICAL) - Code-documentation mismatches
  3. Diataxis Findings (HIGH) - Category misalignments affecting usability
  4. Structure Findings (MEDIUM) - Organizational and navigation issues
  5. Style Findings (LOW) - Style guide compliance

Checkpoint: ✓ Stage 7 complete: Consolidated report generated with findings from [N] stages


Error Handling

If a stage fails to complete, handle as follows:

Build Validation Failure
  • Report build errors immediately
  • STOP the workflow - do not proceed to content analysis
  • Include build errors in the final report
Skill Invocation Errors
  • Report the error in the corresponding report section
  • Continue with remaining stages
  • Note the failure in the Summary section
Missing Dependencies
  • Report what's missing (e.g., "Style guide not found at docs/style-guide.md")
  • Mark that stage as "Incomplete" in the report
  • Continue with other stages
Incomplete Stages

If any stage could not be completed, add an "Incomplete Stages" section to the report listing:

  • Which stage failed
  • Why it failed
  • What's needed to complete it

Constraints

  • Provide criticism and suggestions rather than direct bulk rewrites.
  • Do not modify source code to fix documentation without explicit request.
  • Before restructuring large documentation sections (for example, moving files between tutorial and how-to), ask first.
  • Before suggesting new coverage entities, categories, or metadata patterns, ask first.
  • If code examples seem correct but do not match your understanding of the codebase, ask first.

© canonical, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in .github/skills/documentation-review of canonical/workshop.

  • SKILL.md
  • references/doc-review-report-template.md

Open the folder on GitHubat commit 0b41ee9

Compare with similar skills

Documentation Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Documentation Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Documentation Review this skillcanonical/workshop114—~1.9kAutomated safety check: PassGPL-3.0
React Performanceaffaan-m/ECC276k1 repos~4.5kAutomated safety check: PassMIT
Performance Profileralirezarezvani/claude-skills28k—~684Automated safety check: PassMIT
Agent Performance Optimizerruvnet/ruflo74k2 repos~3.6kAutomated safety check: PassMIT
Handsontable Performance Testinghandsontable/handsontable22k—~3.4kAutomated safety check: PassCustom licence
Performance Managementsickn33/agentic-awesome-skills47k1 repos~4.1kAutomated safety check: PassMIT

Similar skills

  • React Performance

    affaan-m/ECC

    React and Next.js performance optimization patterns adapted from Vercel Engineering's React Best Practices (https://github.com/vercel-labs/agent-skills).

    276k GitHub starsUsed in 1 repo~4.5k tokens
    DevelopmentAuto-check passed
  • Performance Profiler

    alirezarezvani/claude-skills

    Systematic performance profiling for Node.js, Python, and Go applications.

    28k GitHub stars~684 tokensUpdated 1 mo ago
    DevelopmentAuto-check passed
  • Agent skill for performance-optimizer - invoke with $agent-performance-optimizer

    74k GitHub starsUsed in 2 repos~3.6k tokens
    Auto-check passed
  • Handsontable Performance Testing

    handsontable/handsontable

    Guide to Handsontable's performance-tests package: Playwright scenarios measured through CDP traces and compared against golden baselines taken from the develop branch.

    22k GitHub stars~3.4k tokensUpdated today
    Testing & QAAuto-check passed
  • Performance Management

    sickn33/agentic-awesome-skills

    Performance review register: review type, period, employee and reviewer, KPI, OKR and behaviour scores, overall rating, PIP and promotion flags, development plan.

    47k GitHub starsUsed in 1 repo~4.1k tokens
    Business, Finance & HRAuto-check passed
  • Performance

    udecode/plate

    Review performance lanes with GitHub-scale tactics not owned by Vercel React rules: cohort segmentation, repeated-unit budgets, interaction-level INP, memory tagging, degradation contracts, browser…

    17k GitHub stars~2.6k tokensUpdated today
    Auto-check passed

More from canonical/workshop

  • Documentation Build

    canonical/workshop

    Validates documentation builds successfully. An agent skill from canonical/workshop.

    114 GitHub stars~762 tokensUpdated today
    Auto-check passed
  • Documentation Style

    canonical/workshop

    Enforces project documentation style guide compliance for tone, voice, terminology, punctuation, and formatting.

    114 GitHub stars~714 tokensUpdated today
    Auto-check passed
  • Documentation Verify

    canonical/workshop

    Verifies documentation accuracy by cross-referencing claims, CLI commands, API signatures, and configuration against source code.

    114 GitHub stars~932 tokensUpdated today
    Auto-check passed
  • Documentation Diataxis

    canonical/workshop

    Analyzes documentation against Diataxis framework (Tutorial, How-to, Reference, Explanation).

    114 GitHub stars~1.6k tokensUpdated today
    Auto-check passed
  • Documentation Structure

    canonical/workshop

    Validates documentation structural integrity including heading hierarchy, metadata, file naming, navigation, and cross-references.

    114 GitHub stars~704 tokensUpdated today
    Auto-check passed

Questions about Documentation Review

What does Documentation Review do?

Performs comprehensive documentation review including build validation, Diataxis analysis, structure audit, accuracy verification, and style compliance. Documentation Review is an agent skill from canonical/workshop. Performs comprehensive documentation review including build validation, Diataxis analysis, structure audit, accuracy verification, and style compliance.

When should I use Documentation Review?

Documentation Review fits situations like: reviewing documentation changes; auditing documentation quality.

How do I install Documentation Review in Claude Code?

Run `npx skills add canonical/workshop --skill documentation-review -a claude-code`. Or copy the skill folder (.github/skills/documentation-review in canonical/workshop) into .claude/skills/documentation-review in your project. Claude Code loads it when a task matches its description.

How do I install Documentation Review in Codex?

Run `npx skills add canonical/workshop --skill documentation-review -a codex`. Or copy the skill folder (.github/skills/documentation-review in canonical/workshop) into .agents/skills/documentation-review in your project. Codex loads it when a task matches its description.

Can I use Documentation Review in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add canonical/workshop --skill documentation-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/documentation-review, .gemini/skills/documentation-review, .github/skills/documentation-review and .opencode/skills/documentation-review in your project.

What does Documentation Review need to run?

SKILL.md names no scripts, command-line tools or credentials: Documentation Review is instructions for the agent only.

Does Documentation Review access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Documentation Review safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Documentation Review use?

Documentation Review is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Documentation Review use?

About 1.9k tokens (SKILL.md is roughly 7.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 339 tokens, read only when the agent opens those files.

What are the alternatives to Documentation Review?

Skills that share tags, products or a category with Documentation Review: React Performance (affaan-m/ECC, 276k stars), Performance Profiler (alirezarezvani/claude-skills, 28k stars), Agent Performance Optimizer (ruvnet/ruflo, 74k stars) and Handsontable Performance Testing (handsontable/handsontable, 22k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Documentation Review?

canonical (a GitHub organization) maintains it in canonical/workshop, which has 114 GitHub stars. The repository holds 6 skills in this directory. The repository was last updated on October 9, 2026.

Source: canonical/workshop on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.