--- name: google-search-console-new-queries description: "Every Thursday, the Search Console queries that brought impressions this week and none in the 12 weeks before, with the page Google picked for each. Use when the user asks for this, or when it runs on a schedule." --- # Search Console new queries Every Thursday, the Search Console queries that brought impressions this week and none in the 12 weeks before, with the page Google picked for each. New queries are early signals. They show a topic Google has started to connect with you, a page that broke into results for wording you never targeted, or demand that did not exist last quarter. In the Performance report they hide among thousands of familiar rows. The agent compares this week with the 12 weeks before and keeps only what is genuinely new. The list is short enough to read every week, and it often ends in a content idea, or a section to add to a page that is already halfway there. ## Two ways to get the data Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used. ## Route A: API or MCP server ### What you need - Access to the Search Console property as a full or restricted user. - The Search Console API for the agent, from an MCP server or connector, or OAuth credentials with the `webmasters.readonly` scope. If something is missing, say what and stop; do not invent numbers. ### Steps 1. **Pull this week's queries** (Google Search Console): Call `searchanalytics.query` with `dimensions: ["query", "page"]` for the latest 7 days with data (ending 2 days ago) and `rowLimit: 25000`. 2. **Pull the 12 weeks before** (Google Search Console): Query `dimensions: ["query"]` for the 84 days before that window, paging with `startRow` until every row is in. This is the list of queries you already had. 3. **Keep the newcomers**: A query is new when it has at least 20 impressions this week and does not appear in the 84-day list. Merge near-duplicates (plurals, reordered words) so one idea is not listed three times. 4. **Group by theme and page**: Cluster new queries by ranking page and shared words. A cluster with no matching page is a content gap; a query ranking 10 or better on an existing page is a section to add. 5. **List them with a suggestion**: The count and combined impressions, then up to 10 queries with impressions, clicks, position and page, and one suggestion per cluster. ## Route B: the browser ### What you need - Signed in to Google with at least Restricted access to the Search Console property, with a Google account that can create Sheets for the two exports. If something is missing, say what and stop; do not invent numbers. ### Steps Start at https://search.google.com/search-console/performance/search-analytics. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears. 1. **Export this week's queries**: Pick the property, open Performance > Search results, set the Date filter to `Last 7 days` and turn on all four metrics. On the Queries tab, filter the rows to `Impressions` greater than `20` with the filter icon above the table, then Export > Google Sheets. 2. **Export the 12 weeks before**: Change the Date filter to Custom, from 91 days ago to 8 days ago, and export the Queries tab again. This is the list of queries you already had; it is capped at 1,000 rows, so it holds the queries with real volume. 3. **Keep the newcomers**: In the sheet, keep queries from the 7-day export that are absent from the 84-day export. Merge plurals and reordered words so one idea is not listed three times. 4. **Find the page Google picked**: Back in the report with the 7-day range, click each new query and open the Pages tab for its ranking page; remove the filter before the next one. Cluster new queries by page and shared words. 5. **Suggest a page or a section**: A cluster with no matching page is a content gap; a query ranking 10 or better on an existing page is a section to add. Start with the count and combined impressions, then up to 10 queries. Note: The 84-day export is capped at 1,000 rows sorted by clicks, so a rare old query can be missing from it and look new; the API sees the whole long tail. Rows under the 20-impression threshold are not worth the extra clicking either way. ## The report - **New queries this week**: how many, and their combined impressions. - **Top 10** with impressions, clicks, position and the page Google ranks. - **Clusters** of related new queries, each marked content gap (no page yet) or section to add (page exists). - A content idea from the strongest cluster. ## Post it to Briefkin When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on. ```bash [ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \ -H "authorization: Bearer $BRIEFKIN_TOKEN" \ -H "content-type: application/json" \ -d @- <<'JSON' { "v": 1, "event_id": "google-search-console-new-queries/", "item_key": "google-search-console-new-queries", "project": "example.com", "source": "Search Console new queries skill", "kind": "update", "title": "27 new queries, two about e-invoicing", "summary": "27 queries brought impressions this week after none in 12 weeks, **1,940 impressions** in all. Two of the top three are about e-invoicing rules, and /blog/e-invoicing-rules has no section on Peppol yet.", "widget": [ { "type": "Table", "props": { "columns": [ "Query", "Impressions", "Clicks", "Position", "Page" ], "align": [ "left", "right", "right", "right", "left" ], "rows": [ [ "e-invoicing mandate 2027", 640, 12, 9.1, "/blog/e-invoicing-rules" ], [ "invoice template google docs", 410, 9, 14.3, "/templates/invoice-word" ], [ "peppol invoice format", 290, 3, 18.7, "/blog/e-invoicing-rules" ], [ "how to invoice as a contractor", 180, 4, 11.6, "/blog/how-to-write-an-invoice" ], [ "self billing invoice", 120, 0, 23.4, "/guides/vat-invoice-requirements" ] ] } }, { "type": "Badges", "props": { "items": [ { "label": "Content gap: Google Docs template", "tone": "warn" }, { "label": "Section to add: Peppol", "tone": "info" } ] } } ] } JSON ``` Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report. ## Run it on a schedule Suggested: Thursdays at 9:00 (`0 9 * * 4`). From cron, with the agent signed in on that machine: ```bash 0 9 * * 4 claude -p "Run the google-search-console-new-queries skill" # Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run ``` Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/google-search-console-new-queries#schedule Source: https://briefkin.dev/skills/google-search-console-new-queries