seo-geo-answerblocks (M11)

AI engines cite passages they can lift whole, without surrounding context — so passage structure is the highest-leverage on-page AI signal. This module checks extractability; for who is allowed to fetch and cite the page, see references/ai-crawlers.md.

Audits

Working from the PageSnapshot (rendered_dom if present, else raw_html):

Heading → direct answer: for each H2/H3 that reads as a question (or implies one), is the first paragraph a direct, ~40-60 word answer — not a windup ("In this section we'll explore…")?
Self-contained passages: are top-level passages ~134-167 words and answerable on their own? Flag unresolved anaphora ("as mentioned above", "this", "the latter") that needs prior context to parse.
Structured forms: presence of lists, comparison tables, and explicit term definitions where the content warrants them — these are disproportionately quoted.
Question-shaped headings: do headings phrase real user questions (how/what/why/when), matching likely prompts?
TL;DR / summary blocks: an up-front summary or key-takeaways block the engine can extract verbatim.
Semantic completeness: can each passage answer its heading with zero external context — the strongest single correlate of AI citation. This is the check that gates the score.

Fixes

PROPOSED: answer-block rewrites (heading → tight 40-60 word lead) and TL;DR drafts, generated from existing page content, surfaced one item at a time for the user to accept/edit. Each becomes a fix_preview diff only after acceptance.
ADVISORY: structural suggestions ("split this 400-word passage", "add a definition list here") where intent is editorial.
Never silently rewrite published prose, and never invent facts to fill a passage — if a claim is missing leave a clearly-marked TODO placeholder for the user. No finding here is auto.