---
description: Review blog article against editorial style guide using 8 specialized agents
argument-hint: [brand-name] [article-path] [--auto-apply]
---

# Style Check: Multi-Agent Article Review

Review the article at **$2** against the **$1** brand style guide using 8 specialized agents running in parallel.

## Step 1: Validate Inputs

Before proceeding:
- Verify style guide exists at `brands/$1/style-guide.md`
  - If missing: Error and suggest running `/generate-style-guide $1 [source]`
- Verify article exists at $2
  - If missing: Error with clear message
- Check if $3 is `--auto-apply` (determines whether to apply corrections automatically)

## Step 2: Read Required Files

Read the following files:
- Article to review: $2
- Brand style guide: `brands/$1/style-guide.md`
- All 8 agent prompts from `.claude/prompts/`:
  - voice-tone-agent.md
  - grammar-usage-agent.md
  - punctuation-agent.md
  - formatting-agent.md
  - technical-standards-agent.md
  - content-patterns-agent.md
  - industry-terms-agent.md
  - quick-reference-agent.md

## Step 3: Launch 8 Specialized Agents in Parallel

**CRITICAL: Use Task tool to launch ALL 8 agents simultaneously in a SINGLE message.**

Each agent should receive:
- The agent prompt from `.claude/prompts/[agent-name].md`
- The complete article content from $2
- The relevant section(s) from `brands/$1/style-guide.md`

**Launch these 8 agents in parallel:**

1. **Agent 1 - Voice & Tone**: Review against Section 1 of style guide
2. **Agent 2 - Grammar & Usage**: Review against Section 2 of style guide
3. **Agent 3 - Punctuation**: Review against Section 3 of style guide
4. **Agent 4 - Formatting**: Review against Section 4 of style guide
5. **Agent 5 - Technical Standards**: Review against Section 5 of style guide
6. **Agent 6 - Content Patterns**: Review against Section 6 of style guide
7. **Agent 7 - Industry Terms**: Review against Section 7 of style guide
8. **Agent 8 - Quick Reference**: Final comprehensive check against Section 8

**Each agent must return:**
- List of violations with line numbers
- Specific corrections for each issue
- Citation of relevant style guide rules

## Step 4: Aggregate Findings

Once all 8 agents complete:
- Collect all violation reports
- Remove duplicate findings (if multiple agents caught same issue)
- Sort violations by line number (top to bottom)
- Calculate total violations by category
- Determine severity (High/Medium/Low)

## Step 5: Generate Review Report

Create a structured report:

```markdown
# Style Check Report: $1

**Article:** $2
**Brand:** $1
**Review Date:** [current date]
**Total Violations:** [number]

---

## Summary by Section

- Voice & Tone: [X] violations
- Grammar & Usage: [X] violations
- Punctuation: [X] violations
- Formatting: [X] violations
- Technical Standards: [X] violations
- Content Patterns: [X] violations
- Industry Terms: [X] violations
- Additional (Quick Check): [X] violations

---

## Detailed Findings

### Agent 1 - Voice & Tone
[Paste agent findings]

### Agent 2 - Grammar & Usage
[Paste agent findings]

### Agent 3 - Punctuation
[Paste agent findings]

### Agent 4 - Formatting
[Paste agent findings]

### Agent 5 - Technical Standards
[Paste agent findings]

### Agent 6 - Content Patterns
[Paste agent findings]

### Agent 7 - Industry Terms
[Paste agent findings]

### Agent 8 - Quick Reference Check
[Paste agent findings]

---

## Priority Recommendations

**High Priority** (Brand voice/structural issues):
[List critical violations]

**Medium Priority** (Consistency/formatting):
[List standard violations]

**Low Priority** (Minor polish):
[List minor violations]

---

## Next Steps

- Review findings above
- To apply corrections automatically: `/article-edit $1 $2`
- To apply manually: Edit the article at $2 using the corrections provided
```

## Step 6: Conditional Auto-Apply

**If $3 is `--auto-apply`:**
- Proceed to apply all corrections automatically
- Work through violations from top to bottom by line number
- Use the Edit tool to apply each correction
- Save result to same path with `_edited` suffix (unless already in `edited/` folder)
- Generate before/after summary

**If $3 is NOT `--auto-apply`:**
- Display the report
- Prompt user: "Review complete. Run `/article-edit $1 $2` to apply corrections."

## Step 7: Save Report

Save the review report to:
- `docs/logs/[YYYY-MM-DD]_style-check_$1_[article-name].md`

## Usage Examples

```bash
# Review only (no auto-apply)
/style-check animalz brands/animalz/articles/original/my-article.md

# Review and automatically apply corrections
/style-check animalz brands/animalz/articles/original/my-article.md --auto-apply

# Review article from different location
/style-check hubspot /Users/name/Documents/draft-article.md
```

## Error Handling

- **Style guide missing**: Provide helpful error with suggestion to generate one
- **Article not found**: Check path and provide clear error
- **Agent failures**: If any agent fails, continue with others and note the failure
- **No violations found**: Celebrate! Report that article is fully compliant

## Performance Notes

- 8 agents running in parallel: ~5-10 minutes
- Sequential would take: ~40+ minutes
- Speedup factor: 4-5x