Documentation Review

Review documentation changes for compliance with the Metabase writing style guide.

You say
Install this skill Read the source first Free Written by metabase · unverified publisher
Context cost
1.7k tokensestimated from the bundle, loaded when it triggers
Bundle
1 file · 6.8 kBtext throughout, nothing executable
Licence
AGPL-3.0free to use
Last change
no release on file
Servers it uses
Noneruns standalone

What it does

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

Installed, it changes the agent in these ways.

What this skill changes about the agent is not written down here yet. The listing was collected from its source, and the description is in its own SKILL.md.

Voice & style

Changes how the agent writes and reviews.

contentdocumentationwriting

The skill itself

This is the whole product. A skill is instructions the model reads, so there is nothing behind the listing you cannot see first — the front matter loads with every session, and the body below it loads when the skill triggers.

SKILL.md6.8 kB · 164 lines
--- name: docs-review description: Review documentation changes for compliance with the Metabase writing style guide. Use when reviewing pull requests, files, or diffs containing documentation markdown files. allowed-tools: Read, Grep, Bash, Glob ---
7# Documentation Review Skill
8
9@./../_shared/metabase-style-guide.md
10
11## Review mode detection
12
13**IMPORTANT: Before starting the review, determine which mode to use:**
14
151. **PR review mode**: If the mcp__github__create_pending_pull_request_review tool is available, you are reviewing a GitHub PR
16 - Use the pending review workflow to post all issues as one cohesive review
17 - Follow the workflow steps in "PR review mode format" below
18
192. **Local review mode**: If the MCP tool is NOT available, output issues in the conversation
20 - Format all issues in a numbered markdown list (as described in "Feedback format" below)
21
22## Review process
23
241. **Detect review mode** - Check if mcp__github__create_pending_pull_request_review is available
252. Read the changes through once to understand intent
263. Check all issues that violate style guide or significantly impact readability
274. Only flag issues worth mentioning - if it won't make a material difference to the reader, skip it
285. **REQUIRED: Number ALL feedback sequentially** - Start from Issue 1 and increment for each issue found
29
30## Review checklist
31
32Run through the diff looking for these issues:
33
34**Tone and voice:**
35
36- [ ] Formal/corporate language ("utilize" not "use", "offerings", etc.)
37- [ ] "Users" instead of "people" or "companies"
38- [ ] Excessive exclamation points or overly peppy tone
39- [ ] Telling readers something is cool instead of showing them
40
41**Structure and clarity:**
42
43- [ ] Important information buried instead of leading
44- [ ] Verbose text that adds little value
45- [ ] Paragraphs without clear purpose
46- [ ] Vague headings that don't convey the point
47- [ ] Instructions explain "why" before telling "what to do"
48- [ ] Tasks described as "easy" or "simple"
49
50**Links and references:**
51
52- [ ] Linking the word "here" instead of descriptive text
53- [ ] Links in headings (unless entire heading is a link)
54
55**Formatting:**
56
57- [ ] Ampersands as "and" substitute (except proper nouns)
58- [ ] Inconsistent list formatting
59
60**Code and examples:**
61
62- [ ] Code examples that don't work or would error
63- [ ] Commands not in execution order
64- [ ] Full-width screenshots instead of scoped UI elements
65- [ ] Excessive or unnecessary images
66
67**Sentence construction:**
68
69- [ ] Overuse of pronouns when introducing new terms
70
71## Quick scan table
72
73| Pattern | Issue |
74| ----------------------------- | --------------------------------------------- |
75| we can do X, our feature | Should be "Metabase" or "it" |
76| click here, read more here | Need descriptive link text |
77| easy, simple, just | Remove condescending qualifiers |
78| users | Should be "people" or "companies" if possible |
79
80## Feedback format
81
82**MANDATORY REQUIREMENT: Every single issue MUST be numbered sequentially starting from Issue 1.**
83
84This numbered format is NON-NEGOTIABLE. It allows users to efficiently reference specific issues (e.g., "fix issues 1, 3, and 5") and track which feedback has been addressed.
85
86### Local review mode format
87
88When outputting issues in the conversation (local mode), use this format:
89
90```markdown
91## Issues
92
93**Issue 1: [Brief title]**
94Line X: Succinct description of the issue
95[code or example]
96Suggested fix or succinct explanation
97
98**Issue 2: [Brief title]**
99Line Y: Description of the issue
100Suggested fix or explanation
101
102**Issue 3: [Brief title]**
103...
104```
105
106**Examples:**
107
108> **Issue 1: Formal tone**
109> Line 15: This could be more conversational. Consider: "You can't..." instead of "You cannot..."
110
111> **Issue 2: Vague heading**
112> Line 8: The heading could be more specific. Try stating the point directly: "Run migrations before upgrading" vs "Upgrade process"
113
114### PR review mode format
115
116When posting to GitHub (PR mode), use the **pending review workflow**:
117
118**Workflow steps:**
119
1201. **Start a review**: Use mcp__github__create_pending_pull_request_review to begin a pending review
121 - This creates a draft review that won't be visible until submitted
122
1232. **Get diff information**: Use mcp__github__get_pull_request_diff to understand the code changes and line numbers
124 - This helps you determine the correct file paths and line numbers for comments
125
1263. **Identify ALL issues**: Read through all changes and identify every issue worth mentioning
127 - Collect all issues before posting any comments
128 - Number them sequentially (Issue 1, Issue 2, Issue 3, etc.)
129
1304. **Add review comments**: Use mcp__github__add_pull_request_review_comment_to_pending_review for each issue
131 - **CRITICAL**: Post ALL comments in a SINGLE response using multiple tool calls in parallel
132 - Each comment should reference a specific file path and line number from the diff
133 - Start each comment body with **Issue N: [Brief title]**
134 - Include the description and suggested fix
135
1365. **Submit the review**: Use mcp__github__submit_pending_pull_request_review to publish all comments at once
137 - Use event type "COMMENT" (NOT "REQUEST_CHANGES") to make it non-blocking
138 - **Do NOT include a body message** - Leave the body empty or omit it entirely
139 - All comments will appear together as one cohesive review
140
141**Comment format example:**
142
143```
144**Issue 1: Formal tone**
145
146This could be more conversational. Consider: "You can't..." instead of "You cannot..."
147```
148
149**IMPORTANT**:
150- Each issue gets its own review comment attached to the pending review
151- Number ALL comments sequentially (Issue 1, Issue 2, Issue 3, etc.)
152- Always start the comment body with **Issue N: [Brief title]**
153- **MUST add all comments in parallel in a single response** - Do NOT add them one after another in separate responses
154- Do NOT output a summary message to the conversation - only post GitHub review comments
155- When submitting the review, do NOT include a body parameter (or leave it empty) to avoid cluttering the PR with summary text
156- The review will appear as a single review with multiple comments when submitted
157
158## Final check
159
1601. Remove any issues from your assessment that won't make a material difference to the reader if addressed. Only flag issues worth the author's time to fix.
1612. **Verify all issues are numbered sequentially** starting from Issue 1 with no gaps in numbering.
1623. Confirm the format exactly matches: **Issue N: [Brief title]** where N is the issue number.
1634. **In PR mode**: Verify each issue was posted as a separate GitHub comment (not output to conversation).
164
In the file
SKILL.md1,029 words
Files1
LicenceAGPL-3.0
Why you can read it

Nothing in a skill executes. The client loads the text and the model follows it, so a skill can be audited the way a runbook is — by reading it.

What it costs in context

Skills are not billed by the call. They are paid for in context: every token the instructions occupy is a token your code, your diff and your conversation cannot use. Here is what this one takes and when it takes it.

≈60
always loaded
The name and description, so the model knows the skill exists and when to reach for it.
1,640
on trigger
The instruction body, read only when the skill fires.
0.85%
of a 200k window
Ten skills this size would take about 9% of the window before you open a file.
050k100k150k200k context window

1.7k tokens, estimated from the bundle at four bytes to the token, held for the rest of the session once it triggers. Middling. Fine to keep on in a project where you use it weekly, worth unloading in one where you never do.

Servers bill, skills cost

A server charges by the month. A skill charges once per session, in context, and then keeps charging it for as long as the session lives.

Before and after

The same question, put to the same model twice: once as it comes, and once with these instructions loaded.

No worked example has been published for this skill yet.

Adoption
Installsnone yet
Ratingno reviews yet

The procedure it runs

The procedure has not been published here. It is in the skill’s own SKILL.md, which its author has not sent to the marketplace yet.

Prose, not code

These steps are written for a model to follow, not executed by a runtime. It can still be told to skip one, and it will say so when it does.

Servers it uses

None. This skill calls no MCP servers at all.

Everything it needs is in the instructions, so it works in a project with nothing connected — the model reads the file and changes how it works with what it can already reach.

It writes no files and reaches no network. All it changes is how the model reasons and writes.

What it asks for
Writes filesno
Network accessno

Read from the allowed-tools line of this skill’s own SKILL.md. A skill grants no permissions of its own — it can only ask for tools your client already has.

What it will not do

Every skill is narrow, and the useful ones say where they stop. These are the jobs this one is the wrong tool for.

What this skill is not for has not been published here. Nothing is implied by that: it is a section the author has not filled in.

What is in the bundle

1 file, 6.8 kB on disk. A bundle is text throughout: the instructions the model reads, plus the templates it fills in.

  • SKILL.md6.8 kB
What is not in it

No dependencies and nothing executable: a skill is text the agent reads, so the bundle is 1 file you can review in full before installing. The AGPL-3.0 licence covers the templates and examples as well as the instructions.

Install

Installing copies the bundle into your project. Nothing runs at install time — the files sit on disk until the model reads them.

# Documentation Review · 1.7k tokens when loaded npx mcprush@latest skill add metabase/documentation-review

Writes to .claude/skills/documentation-review/ in the current project. Add --global to put it in your home directory instead, for every project.

Which clients pick it up on their own

A skill is a folder of text. A client with a skills folder reads it without being told; everywhere else the same text works, it is just handed to the model rather than found.

Claude Code.claude/skills/
Claude Desktop
ChatGPT
Cursor.cursor/skills/
VS Code.github/skills/
Codex CLI.agents/skills/
Gemini CLI.gemini/skills/
Grok.grok/skills/
Zed.agents/skills/
Windsurf.windsurf/skills/
Agent SDK.claude/skills/
HTTP / API
This release
Versionnot versioned
Publishedno release date on file
PriceFree
Referencemetabase/documentation-review

Versions

Its author publishes no version number, so there is nothing here to pin to: what you install is the folder as it stands today. Instructions change more often than APIs do — a skill can be rewritten entirely without anything it depends on moving.

v
  • No earlier releases have been published to the marketplace.
Pinning

Nothing to pin to: this skill carries no version number of its own. What you install is what the folder holds on the day you install it.

Reviews

no reviews yet · no installs yet

Nobody has reviewed this skill yet. The rating is the mean of the reviews written here, so there is none until somebody writes the first.

Who can post

Only accounts that have had the skill installed for fourteen days, so a review is written after living with it rather than after reading it. Publishers may reply once.

Publisher
Servers0
Claim this skill