← ClaudeAtlas

search-consolelisted

Pull live Google Search Console data through its API with the user's own access, without exports, for clicks, impressions, click-through and average position by query, page, country, device or date; URL inspection (whether each page is indexed, why not, Google's chosen canonical, last crawl, robots.txt, rich result issues); and submitted sitemaps with their errors. Use when someone asks for Search Console data, Google rankings or positions, indexing status, why a page is not indexed, index coverage, crawl dates, canonical problems, sitemap errors, or wants keyword research or SEO reporting from live data instead of a CSV export.
visibian/seo-geo-skills · ★ 0 · AI & Automation · score 75
Install: claude install-skill visibian/seo-geo-skills
# Search Console, live Read the site's own Google Search Console data through the API, so the other skills work from today's numbers instead of an export. ## Before starting The user needs access to the property in Search Console and an OAuth access token for it, set in their environment as `SEARCH_CONSOLE_TOKEN`. Explain the two ways in the script's help (`python3 scripts/gsc_api.py --help`): Google's OAuth 2.0 Playground (no setup; the token lasts one hour) or gcloud. The scope is `webmasters.readonly`: the token can read, not change, anything. Never ask the user to paste a token, key or password into the chat. If they cannot set a token, use the exports instead: the `keyword-research` skill reads the Performance export, and Search Console's own URL Inspection tool shows one page at a time. ## Steps From this skill's folder: 1. Find the property's exact name: `python3 scripts/gsc_api.py sites`. A domain property reads `sc-domain:example.com`; a URL-prefix property is the full address with its trailing slash. 2. Performance: `python3 scripts/gsc_api.py performance sc-domain:example.com --days 90` writes `search-console-performance.csv` with query, page, clicks, impressions, ctr and position. Then run the `keyword-research` skill on it (`python3 gsc_queries.py search-console-performance.csv --brand "..."`), which adds the pages that compete for one query. Other cuts: `--dimensions query,country`, `--dimensions date` for a trend, `--type image` or `news` or `discover