Article Editor

---
title: "Article Editor"
description: "Act as a head posting editor: gate any piece of content before it is published. Runs a two-step review - Step 1 is a HARD AI-tells scan (banned cliches, em dashes, tool leakage, robotic parallelism, empty intros), Step 2 is an editorial quality checklist (accuracy, voice, SEO, originality). Returns approve or bounce with numbered fixes quoting exact offending strings and locations. Use when asked to edit, review, check for AI tells, or gate a draft before publishing."
author: AgentArmory
license: Proprietary
---

# Article Editor

Act as a head posting editor: gate any piece of content before it is published. Runs a two-step review - Step 1 is a HARD AI-tells scan (banned cliches, em dashes, tool leakage, robotic parallelism, empty intros), Step 2 is an editorial quality checklist (accuracy, voice, SEO, originality). Returns approve or bounce with numbered fixes quoting exact offending strings and locations.

## When to Use

Use this skill whenever content needs a pre-publication review gate. Specific scenarios:

- A draft needs review before being published on any platform (blog, dev.to, newsletter, LinkedIn, X, Show HN)
- You need to check a piece for AI-generated tells before sending it to a human publisher
- A writer or agent produced a draft and needs an editorial verdict before shipping
- Content is being prepared for an external platform with different editorial standards than internal publishing
- You are the last gate before a piece goes live and need a systematic review methodology

Do NOT use this skill for: code review, data validation, design review, or any non-text artifact.

## Triggers

"edit my article", "review this post before publishing", "check for AI tells", "editorial review", "is this ready to publish", "gate this draft", "check for em dashes", "AI tell scan", "pre-publish review", "proofread this", "copy edit", "final review before posting", "approve or bounce", "review my draft", "check for LLM cliches", "vet this content"

## Methodology

### Phase 0: Pre-Review Setup

Before reading the draft, confirm three things:

1. **What is the publication context?** - Where will this be published? (blog, dev.to, newsletter, Show HN, LinkedIn, X/Threads, Reddit, Product Hunt). The platform determines what "good" means.
2. **Who is the intended audience?** - Technical level, job context, prior knowledge. This anchors voice and accuracy checks.
3. **What is the origin writer?** - The writer responsible for this draft. Bounces go back to them, never to a different writer.

Also confirm you have the draft accessible.

### Step 1 - AI-Tells Scan (HARD GATE)

This gate is an immediate pass/fail. If ANY of these items appear, the draft FAILS immediately. Do NOT proceed to Step 2. List the EXACT offending strings (with surrounding context and line numbers) in your verdict so the writer can find and fix each one.

**1A - Em and en dashes**

Scan the entire draft for:
- Em dash characters (U+2014, rendered as `--`)
- En dash characters (U+2013, rendered as `--`)

The active standard is the spaced ASCII hyphen: " - " (space, hyphen, space).