← ClaudeAtlas

competitor-analyzerlisted

Read at least three competitors' actual content before a word gets written, and turn it into a gap analysis and a word count to beat. Pulls the live SERP, reads the real pages (not just their headings), and reports what they all cover, what to go deeper on, what nobody covers, and the target length. Pulls real People Also Ask from the SERP, never invented. Run after keyword-researcher, before outline-drafter. Produces a gap analysis and stops.
robdasi/skills · ★ 0 · Data & Documents · score 75
Install: claude install-skill robdasi/skills
# Competitor Analyzer This is the number one rule, the one that's skipped most and costs most: never write an article without reading at least three of the competitors actually ranking for the term. Not their headings, their real content. Every time it's skipped, the article comes out short, with obvious gaps the ranking pages already filled. So this skill reads them first, and turns them into the bar the new page has to clear and the gap where it can win. Run after `keyword-researcher`, before `outline-drafter`. ## Inputs (ask for whatever is missing) - **The keyword brief** from `keyword-researcher` (required). - **Search access**: a live SERP source, or web search as the fallback. Real search first, always. ## The method 1. **Pull the live SERP and read the top three.** Not the headings, the actual content of at least the top three ranking pages. If a page ranks, it counts, even if it's a service page rather than a blog post. What ranks is the standard. 2. **Get the real word count to beat.** Measure the actual length of the top three and take the average. The new article meets or exceeds it. "Their page doesn't really count" is not a reason to come in short. 3. **Write the gap analysis, before any outline.** Four parts: - **Table stakes**: what they all cover. The new page has to cover these too or it looks thin. - **Go deeper**: what they touch but treat shallowly. The room to be better. - **Nobody covers**: the angle, question, or example none of them