Installs into .claude/skills of the current project.
Are you the author of Review Article?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/tomzx-review-article)
---
name: review-article
description: Review an article for accuracy, clarity, structure, sourcing, audience fit, cross-article consistency, and style.
---
# Review Article
Audits an article and reports findings across seven categories: accuracy, clarity, structure, sourcing, audience fit, cross-article consistency, and style.
Each finding is prioritized with π΄ MUST fix, π‘ SHOULD fix, or π’ MAY fix.
## Prerequisites
- Article provided in context or as a file path to read
- Target audience (if known, improves audience-fit check)
- Relevant sources (optional, improves accuracy check)
- Other related articles in the same series (optional, enables cross-article consistency checks)
## Steps
1. Read the article thoroughly.
2. Summarize the message the article is trying to convey in two to three sentences, as you understood it.
This serves as a comprehension check: if the summary diverges from what the author intended, the clarity or structure sections will explain why.
3. Cross-reference claims against provided sources if available.
4. Compare claims, figures, and terminology against other related articles in context if available.
5. Identify issues in each category below.
6. Prioritize each finding: π΄ MUST, π‘ SHOULD, π’ MAY.
7. Report findings using the output format. Omit categories with no findings.
## Review Checklist
### Accuracy
- Are factual claims correct and verifiable?
- Are statistics, numbers, and dates accurate?
- Are technical concepts described correctly?
- Are attributed claims accurately representing their sources?
### Clarity
- Does the article open with the most important point, not background?
- Is every sentence necessary, with no filler or padding?
- Are abstract points anchored with concrete examples, numbers, or comparisons?
- Is jargon avoided or defined on first use?
- Is the language plain: common words over less frequent synonyms, short sentences, active voice, and no figurative language (metaphors, idioms) that the reader must decode?
### Structure
- Does the article follow a logical progression from hook to takeaway?
- Are sections focused on a single idea each?
- Is there a clear through-line connecting all sections?
- Does the ending provide something actionable or memorable (not a weak summary)?
### Sourcing
- Are claims attributed to sources with hyperlinks, not just textual citations?
- Are all hyperlinks verified (no 404s or broken references)?
- Are opinions clearly distinguished from sourced facts?
- Are sources credible and relevant?
### Audience Fit
- Is the tone appropriate for the stated or implied audience?
- Is the depth calibrated: neither too shallow nor too detailed?
- Does the article assume appropriate prior knowledge?
- Would the target audience find this valuable?
### Cross-Article Consistency
- When other related articles are in context, are facts, statistics, and dates consistent across them?
- Is terminology and naming used consistently across related articles?
- Do claims or conclusions in one article contradict another in the series?
- Are cross-article references (links to companion articles) accurate and up to date?
### Style
- Is each sentence on its own line in the markdown source?
- Are em-dashes avoided in favor of commas or parentheses?
- Is the writing concise without hedging every claim?
- Is the same point restated across multiple sections (redundancy)?
- Apply the shared catalog in [`../communication-guidelines/references/writing-rules.md`](../communication-guidelines/references/writing-rules.md), and report each finding by rule ID (for example `writing-rules.md#8` for an em dash).
## Output Format
```markdown
## Summary
π΄ / π’ <Overall assessment in one sentence.>
## Message Summary
<Two to three sentences capturing the message the article conveys, as you understood it.
The author should verify this matches their intent; a mismatch indicates a clarity or structure problem.>
## Accuracy
<Findings with π΄/π‘/π’ priority, or "No issues found.">
## Clarity
<Findings or "No issues found.">
## Structure
<Findings or "No issues found.">
## Sourcing
<Findings or "No issues found.">
## Audience Fit
<Findings or "No issues found.">
## Cross-Article Consistency
<Findings or "No issues found.">
## Style
<Findings or "No issues found.">
```
## Outcome
If `$OUTCOME_YAML` is set, emit your verdict there per `skills/sdlc/references/shared.md`:
| Verdict | When |
|---|---|
| `approved` | No blocking findings; the subject passes review |
| `changes-requested` | Findings the author must address before it passes |
| `rejected` | Fundamental flaw requiring rework or stopping |
## Example Usage
**Scenario 1: Broken hyperlink**
The article cites a paper with a URL that returns 404.
π΄ MUST fix the link or find an alternative source.
**Scenario 2: Padding in the opening**
The article starts with "In today's fast-paced world of technology..." before reaching the point.
π‘ SHOULD rewrite the opening to lead with the core insight.
**Scenario 3: Em-dash usage**
Three sentences use em-dash constructions where commas would work.
π’ MAY replace em-dashes with commas or parentheses.
**Scenario 4: Conflicting figures across articles**
A companion article in the series reports 40% adoption while this one cites 35% for the same metric and period.
π΄ MUST reconcile the discrepancy and align both articles.
**Scenario 5: Not plain language**
A paragraph says the tool "facilitates stakeholders in the utilization of the mechanism," a passive and jargon-heavy way to say it lets people use the feature.
π‘ SHOULD rewrite in plain language: common words, short sentences, active voice.
**Scenario 6: Figurative language**
The article says the migration "moved to the center of the roadmap" and the bottleneck "sheds light on the real problem" instead of stating plainly what changed and what the bottleneck revealed.
π‘ SHOULD replace the metaphors with the literal statement.
## Next Step
Once all π΄ MUST findings are resolved, the article is ready for publication.