agentleFS
UXMagicaiUXMagicai/seosorted-skills/skills/cannibalization-check/SKILL.md

cannibalization-check

Skill0 starsChanged 39 days agoMITView on GitHub

Detect keyword cannibalization by comparing Google's live top-10 results with SEOsorted, before writing new content or when two pages compete. Use when the user asks about cannibalization, duplicate content, two pages competing for the same keyword, whether to merge articles, or before generating a batch of new articles.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Cannibalization check

Two keywords that look different can be one query to Google ("mounjaro weight loss" vs "zepbound weight loss" are not; "metformin side effects in women" vs "metformin side effects" are). Word overlap cannot tell you which. Google's results can: if two queries return mostly the same pages, Google treats them as the same intent, and two articles targeting them will compete.

Steps

  1. Collect the candidate keywords (up to 25 per call).
  2. check_keyword_overlap with terms and compare_existing: true so every term is also compared with the workspace's existing articles. Each term costs one keyword-lookup credit; say so for large lists.
  3. Read the verdicts:
    • 7–10 shared URLs: same intent. Do not write both. Keep the higher-volume keyword; fold the other in as a secondary keyword or a section.
    • 4–6 shared: borderline. Both can exist only with clearly different angles (a dosage chart vs a mechanism explainer). Write the angle into the brief.
    • 0–3 shared: safe.
  4. Conflicts with existing articles: prefer improving the existing page (content-refresh) over writing a new one, unless the existing page targets a different intent.

Output

Table of conflicts: keyword A, keyword B (or existing article), shared URLs, verdict, action. Then the list of terms that are clear to write.

Guardrails

Results reflect today's SERP in the workspace's market. If a fetch fails for a term, report it as unchecked rather than safe.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.