Use your Search Console data to decide which pages and queries need attention.
a-seo-gsc fetches Performance data through the Google Search Console API or reads manual exports. It produces a prioritized action plan with the evidence attached to each recommendation.
The analysis checks impressions, clicks and position together before proposing a fix. Low clicks can have several causes, so a title rewrite is a hypothesis to test. API data includes page-and-query pairs, which show when several of your pages appear for the same query. Manual exports remain supported, but without pairs the query-split check is omitted. Downloaded indexing reports add evidence about pages Google has discovered or crawled but has not indexed.
The arithmetic is done by a script, not by a model reading a CSV. parse_gsc.py pre-processes the export deterministically, so every number in the plan came from counting rather than from a language model's impression of a spreadsheet. That is the difference between a plan you can act on and a plan that sounds right.
It adapts to the site. A 300-page blog, a 50,000-SKU store, a documentation site and a local business fail in different ways, and a checklist written for one of them wastes your time on the others.
The reference doc has the exact click-paths for getting every report out of Search Console, including how to pull two date ranges for trend comparison, and which alternatives work if you do not have GSC access (Bing Webmaster Tools, an Ahrefs or Semrush export).
The fetcher saves raw exports locally and sends requests to Google OAuth and Search Console endpoints. The parser computes findings on your machine; Claude then reads them in your session under your Claude Code plan data controls. The plugin adds no telemetry, and the service-account key stays on your machine.
Ships in the search-visibility plugin, alongside a-geo-optimizer. Run /a-seo-gsc when you want to analyze a property or a set of exports.
/plugin marketplace add imisic/claude-marketplace
/plugin install search-visibility@imisicThen the command is /a-seo-gsc.
Everything mine on this shelf comes from my own working setup, shared for learning. Test it and adapt it to your project before relying on it; you run it at your own risk.