Aeo Content Skills — AI skill for Claude Code

AEO Content Skills

Two agent skills for writing content that ranks on Google **and** gets cited by AI Overviews, ChatGPT, Perplexity, and Claude.

Compatible with Claude Code, Claude Cowork, Codex, Cursor, and other [agentskills.io](https://agentskills.io)-compatible agents.

Skill	Input	Output
content-brief	A keyword	A brief a writer or an agent can execute against
pre-publish-check	A finished draft	Pass/fail on five checks, with specific fixes

Why these exist

The failure mode of AI-assisted content isn't bad writing. It's *competent, well-structured, entirely redundant* writing — pages that read fine and add nothing to what already ranks. Those pages don't rank, don't get cited, and cost real money to produce.

These two skills are built around gates that catch that before it becomes a draft, and before a draft becomes a published page.

---

`content-brief`

Takes a keyword through seven steps:

1. Cannibalization check — do you already have a page targeting this? Two pages fighting over one query is the most common self-inflicted SEO wound, and it's invisible for months.
2. SERP research — the question set people actually search, a coverage map of the top 5, and the content type the SERP is asking for (ideas listicle, commercial roundup, ranked list, explainer).
3. Confirmation gate — you review and correct the research before any structure gets built.
4. The answer block — the two-paragraph, ~120–160-word intro that AI Overviews extract, with keyword placement planned. Written out, not described.
5. The structure — a blueprint for the content type, with search questions mapped to H2s, deduped and intent-filtered. Every section has to stand alone, because LLMs cite passages, not pages. Each section gets direction (named items, not "cover best practices"), a word budget, and a citation slot; the count in the title has to match the count in the outline.
6. The edge gate — see belo