Content refresh
SEOmatic is the SEO agent that works from your real Google Search Console data and makes the fixes on your site. Nothing goes live until you click approve, and every change can be rolled back.
Refreshing a page that already ranked is usually faster than writing a new one: it has links, history and proof it can rank. This skill picks the pages worth it and says exactly what each needs.
Prerequisite
Search Console must be connected. If a tool reports it is not, give the user this link and stop: https://app.seomatic.ai/connect/gsc - it is one browser step, then re-ask. Do not substitute guesses for GSC data.
1. Find the decay
gsc_performance action compare_periods on pages, with a long window
(90 days against the 90 before). Keep pages that lost at least 25% of clicks
AND still have meaningful impressions. A page with no impressions left is a
rewrite or a prune, not a refresh.
gsc_insights action seasonality to remove pages that dip every year.
2. Find what slipped
For each candidate, gsc_performance action dimension_breakdown with
dimension query and filterPage: which queries lost position, and which
new queries appeared that the page barely answers.
3. Compare with what ranks now
serp_competitors action features on the page's main queries, and
site_pages action crawl on the page itself. Look for:
- Out-of-date facts, prices, years, screenshots or steps.
- Sub-questions the ranking query now implies that the page never answers.
- A format mismatch (results now show lists or tools, the page is an essay).
- Thin sections against queries that grew.
keyword_research action metrics on any new query to size it. Use
site_pages action analyze for the page's score and critical issues.
4. The refresh brief
Per page: what lost (queries, positions, clicks with dates), what to add, what to update, what to cut, and the expected outcome as a range. Specific: "Add a section on 2026 pricing; the query 'X pricing' grew to 1,900 impressions and the page never mentions price" is a brief. "Update content" is not.
Pages with nothing worth saving: say so. content_prune exists for them.
Offer to stage the work
Staging needs a paid SEOmatic plan (the key carries the agents:act scope).
It costs no credits, runs nothing, and always waits for the user's approval.
On a free key the staging actions are locked and say which plan includes
them: describe the specific fix instead so the user can pass it on.
Check task_manage action list first so you do not stage a duplicate.
Then task_manage action create with up to 5 tasks: content_refresh per page (pass the decayed queries), freshness_stamp where only the date and facts are stale, content_prune only with the user's sign-off. Each task
needs targetKey (the full page URL, or a topic for content work), a
title that states the goal in the user's words ("Get 'pricing calculator'
onto page 1", never "Optimize meta tags"), and a rationale with the
numbers behind it. Quote cost from each task's estimatedCost.label.
- Tell the user the cards are staged and where to approve them.
- Never approve on the user's behalf. Approval happens in SEOmatic, or
through
task_manageactiondecideonly after the user says yes to that specific task. - Approved auto-executable tasks start within seconds: read the result back
with
task_manageactiongetbefore you claim anything changed.
What not to do
- Do not refresh a page whose only problem is seasonality.
- Do not change a page that is still growing.
- Do not invent volumes. Every number comes from a tool result.