Every Wednesday, the queries where you rank 8 to 20 with real impressions, the page that ranks, and the clicks each could add near the top.
Reads
You get
The card on the right is a sample. Yours shows your own numbers and updates in place every run.
What lands on your board
38 queries rank between 8 and 20 with 200+ impressions over 28 days. The five below could add about 970 clicks a month at the site's usual top-5 CTR of 6.5%.
Striking distance is the cheapest SEO there is. Google already finds the page relevant and ranks it at the bottom of page one or the top of page two, so a sharper title, a section that answers the query directly or two internal links can be enough. Finding the candidates means filtering a Performance export by position, which few people do twice.
The agent does it weekly, scores each query by the clicks it would gain near the top, and names the page to edit. The list shifts a little every week, which is how you can tell the edits worked.
Call searchanalytics.query with dimensions: ["query", "page"], the 28 days ending 2 days ago, type: "web" and rowLimit: 25000, paging with startRow if the site has more rows.
Filter to rows with an average position from 8 to 20 and at least 200 impressions. Drop branded queries (your name and its misspellings), which are not won with content.
Take the site's own CTR for queries in positions 3 to 5 over the same 28 days, multiply it by each query's impressions and subtract the clicks it already gets. Sort by that gain.
When several candidates point at the same page, list them together, since one rewrite can lift all of them. A query that ranks with two of your pages is a cannibalization hint; note it.
The top 5 to 10 queries with page, position, impressions, clicks and estimated gain, plus one line naming the edit to make first.
With BRIEFKIN_TOKEN set, the agent sends the result as one card that updates in place every run. Without it, you get the report in the chat.
Pick your agent, where the data comes from and how to schedule it; the commands below follow.
One command. It puts the SKILL.md in ~/.claude/skills and in ~/.agents/skills, which Codex, Cursor, Gemini CLI, Copilot CLI and OpenCode read too.
shellcurl -fsSL https://briefkin.dev/skills/google-search-console-striking-distance-keywords/install | shAsk Claude Code:
promptRun the google-search-console-striking-distance-keywords skillOr paste the prompt instead of installing:
promptUsing the Google Search Console API (searchanalytics.query, dimensions query and page, last 28 days ending 2 days ago, up to 25,000 rows), find queries with an average position between 8 and 20 and at least 200 impressions, excluding my brand name. Work out my site's CTR for positions 3 to 5 over the same period and estimate the extra clicks each query would get there (impressions times that CTR, minus current clicks). List the top 10 by that gain with query, page, position, impressions and clicks. Group queries that share a page, flag any query where two of my pages rank, and tell me which page to edit first and what to change.Wednesdays at 9:00, from cron on any machine where Claude Code is signed in:
crontab0 9 * * 3 claude -p "Run the google-search-console-striking-distance-keywords skill"Add it with crontab -e. A laptop asleep at that time runs it when it wakes if you use launchd or Task Scheduler instead; GitHub Actions works for the API route.
Connect Briefkin once with curl -fsSL https://briefkin.dev/install | sh and every run lands on your board as a report card. How setup works
SKILL.md---
name: google-search-console-striking-distance-keywords
description: "Every Wednesday, the queries where you rank 8 to 20 with real impressions, the page that ranks, and the clicks each could add near the top. Use when the user asks for this, or when it runs on a schedule."
---
# Search Console striking distance keywords
Every Wednesday, the queries where you rank 8 to 20 with real impressions, the page that ranks, and the clicks each could add near the top.
Striking distance is the cheapest SEO there is. Google already finds the page relevant and ranks it at the bottom of page one or the top of page two, so a sharper title, a section that answers the query directly or two internal links can be enough. Finding the candidates means filtering a Performance export by position, which few people do twice.
The agent does it weekly, scores each query by the clicks it would gain near the top, and names the page to edit. The list shifts a little every week, which is how you can tell the edits worked.
## 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
- Restricted or full user access to the Search Console property.
- Search Console API access for the agent: a Search Console MCP server or connector, or OAuth or service-account credentials with the `webmasters.readonly` scope.
If something is missing, say what and stop; do not invent numbers.
### Steps
1. **Pull query and page pairs** (Google Search Console): Call `searchanalytics.query` with `dimensions: ["query", "page"]`, the 28 days ending 2 days ago, `type: "web"` and `rowLimit: 25000`, paging with `startRow` if the site has more rows.
2. **Keep positions 8 to 20**: Filter to rows with an average `position` from 8 to 20 and at least 200 impressions. Drop branded queries (your name and its misspellings), which are not won with content.
3. **Estimate the upside**: Take the site's own CTR for queries in positions 3 to 5 over the same 28 days, multiply it by each query's impressions and subtract the clicks it already gets. Sort by that gain.
4. **Group by page**: When several candidates point at the same page, list them together, since one rewrite can lift all of them. A query that ranks with two of your pages is a cannibalization hint; note it.
5. **Post the shortlist**: The top 5 to 10 queries with page, position, impressions, clicks and estimated gain, plus one line naming the edit to make first.
## Route B: the browser
### What you need
- Signed in to Google with at least Restricted access to the Search Console property.
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. **Open 28 days of query data**: Pick the property, then Performance > Search results. Set the Date filter to `Last 28 days`, Search type `Web`, and turn on Total clicks, Total impressions, Average CTR and Average position above the chart.
2. **Keep positions 8 to 20 with real impressions**: On the Queries tab, add a filter `Query > Queries not containing` your brand name. Use the filter icon above the table to filter rows by `Impressions` greater than `200`, then sort by Position and read the rows between 8 and 20.
3. **Read your top-5 CTR**: Change the row filter to `Position` smaller than `5.5` and note the CTR of queries sitting at positions 3 to 5. That is the CTR each candidate would earn near the top.
4. **Find the page for each candidate**: Click a query row, open the Pages tab and note the first page; two of your pages sharing the impressions is a cannibalization hint. Remove the query filter before the next one.
5. **Score and shortlist**: Estimated gain is impressions times your top-5 CTR minus current clicks. Group candidates by page and list the top 5 to 10. When the table hits its 1,000-row cap, use Export > Google Sheets and do the filtering there.
Note: The Queries table shows at most 1,000 rows and only one row filter at a time, so on a large site the position filter leaves out candidates; the export to Google Sheets is the honest version. The UI shows one dimension at a time, so query and page pairs take one click per query.
## The report
- **How many queries** sit in positions 8 to 20 with 200+ impressions, and the clicks they could add.
- **Shortlist**: query, ranking page, position, impressions, current clicks and estimated gain.
- **Pages with several candidates**, since one edit can lift them all.
- **Cannibalization hints**: queries where two of your pages split the impressions.
- The edit to make first.
## 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-striking-distance-keywords/<date of this run, e.g. 2026-10-16>",
"item_key": "google-search-console-striking-distance-keywords",
"project": "example.com",
"source": "Search Console striking distance keywords skill",
"kind": "update",
"title": "38 queries one push from the top 5",
"summary": "38 queries rank between 8 and 20 with 200+ impressions over 28 days. The five below could add about **970 clicks** a month at the site's usual top-5 CTR of 6.5%.",
"widget": [
{
"type": "Table",
"props": {
"columns": [
"Query",
"Page",
"Position",
"Impressions",
"Clicks",
"Est. gain"
],
"align": [
"left",
"left",
"right",
"right",
"right",
"right"
],
"rows": [
[
"invoice template word",
"/templates/invoice-word",
8.4,
6200,
105,
"+298"
],
[
"how to write an invoice",
"/blog/how-to-write-an-invoice",
11.2,
4800,
38,
"+274"
],
[
"invoice number format",
"/blog/invoice-numbering",
9.6,
2900,
41,
"+148"
],
[
"freelance invoice",
"/templates/freelance-invoice",
13.5,
2400,
14,
"+142"
],
[
"proforma invoice meaning",
"/blog/proforma-invoice",
16.1,
1700,
6,
"+105"
]
]
}
},
{
"type": "Callout",
"props": {
"text": "Start with /templates/invoice-word: it ranks 8.4 for \"invoice template word\" and its title never mentions Word.",
"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: Wednesdays at 9:00 (`0 9 * * 3`). From cron, with the agent signed in on that machine:
```bash
0 9 * * 3 claude -p "Run the google-search-console-striking-distance-keywords 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-striking-distance-keywords#schedule
Source: https://briefkin.dev/skills/google-search-console-striking-distance-keywords
Restricted or full user access to the Search Console property.
Search Console API access for the agent: a Search Console MCP server or connector, or OAuth or service-account credentials with the webmasters.readonly scope.
GOOGLE_APPLICATION_CREDENTIALS at the key and set GSC_SITE_URL to sc-domain:example.com or https://www.example.com/.webmasters.readonly scope and a stored refresh token works for scheduled runs.Ask for webmasters.readonly and add the service account as a Restricted user; the reports only read.
Positions 8 to 20 bounce around from day to day, and one week of impressions on a mid-volume query is too thin to rank candidates by. Four weeks smooths that out and still reflects recent edits.
Google leaves out queries it considers too rare to show for privacy, so query-level data always totals less than the property. Striking distance is about queries with real volume, which are the ones Google does return.
No. The skill runs with your own agent and gives you the report in the chat. With a Briefkin token set, each run also posts it to your board as a report card, so you can glance at it instead of asking again.
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, Windsurf and Amp all read SKILL.md files. The install command puts it in ~/.claude/skills and ~/.agents/skills, where they look. ChatGPT, the Claude app and Grok take the prompt instead. Pick your agent under "Install and run it" and the commands change to match.
Yes. Switch the data source to "Your browser" and the steps change to the Google Search Console pages themselves: the agent opens search.google.com/search-console/performance/search-analytics in a browser you are signed in to and reads the numbers off the page. Claude Code, Codex, Cursor, Claude app and ChatGPT can drive a browser out of the box; Gemini CLI, Copilot CLI and OpenCode need a Playwright or Chrome DevTools MCP server first. It is slower than the API and needs you signed in, so it suits a report you run by hand or from a scheduler on your own computer.
Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.
Free during the open beta. No card.