Internal Linking
Internal links are the cheapest ranking lever you own, and the hard part is knowing which page should link where. This agent reads each target page to learn what it is actually about, asks Serpstat which OTHER pages of the same domain rank for that topic, then opens the best candidates to check whether the link already exists, where the anchor would fit, and whether the page is even in the same language. It works on pages that do not rank yet — the topic comes from the page's own text, not from its rankings.
@internallinkingHow it works
Read the target
fetches each target page and extracts the terms that describe its topic
Find candidates
asks Serpstat which same-domain pages rank for those terms, and adds the pages the target already links to
Rank
scores each candidate on topical coverage, page authority and language match
Verify
opens the shortlisted donors to check for an existing link, pick the placement sentence, and confirm the language
Write anchors
proposes anchor text in the DONOR page's language, and a priority per recommendation
Parameters
domainstringRequiredThe site to search for donor pages (e.g. example.com)countryselect (231 options)RequiredWhich country's Google data to search — it decides which pages qualify as donors Default: g_ustarget_urlstextRequiredThe pages that should RECEIVE the new links — up to 25 per rundonors_per_targetnumberOptionalMost donor pages to recommend per target (1-50) Default: 5Example queries
Example output
TARGET /blog/what-is-seo-basics/
critical /blog/seo-for-beginners/ anchor: "what is SEO" already linked: no
"Before touching anything technical, it helps to know what is SEO and what it is not."
high /blog/seo-analysis-content/ anchor: "basics of SEO" already linked: no
"Most content audits start from the basics of SEO rather than the tooling."
medium /page-audit/ anchor: "website promotion" already linked: nofollow
TARGET /page-audit/
already covered — 5 related page(s) already link here and were not repeated.
Summary: 2 requested page(s) → 3 recommendation(s) over 3 donor(s); 1 page already coveredOutput formats
One row per recommendation, plus a row for every requested URL that got none and why
source_urlsource_titletarget_urlstatusanchor_textmatched_keywordsscoredonor_positionalready_linkedexisting_anchorexisting_relplacement_quotepriority