Answer product-owner round 1: 0 🔴, 1 🟡; name the five-of-seven bar and what the other answer drops

docs/decisions.md

@@ -258,7 +258,7 @@ Their CommonMark suite exceptions are `node-model` (§The CommonMark suite check
 
 A reader panel on 2026-10-18 read the title again: seven readers across the bot, the LLM pipeline,
 a support agent and a product manager. Two picked a caption, two the picture alone, and three neither,
-each describing hover text, which ADF holds no node for. No output reached five of seven, so the
+each describing hover text, which ADF holds no node for. No output reached the five of seven that settles a panel, so the
 title still reads as the image's caption until `todo.md` item 121 is answered.
 
 ## A mark skips a node that cannot hold it, and a directive mark wrapping only such nodes is refused, as is a `[text](url)` link unless its text is hard breaks alone

todo.md

@@ -77,7 +77,8 @@ chosen by Goal 6 over Goal 7 (§An image reads as an image only alone in its par
 
 Candidate entry: an image's title becomes its caption when the image stands alone in its paragraph,
 because ADF holds no hover text and the caption keeps the title's words. This image keeps "The moon, at night"
-as visible text under the picture.
+as visible text under the picture. The other answer drops the title, so the picture stands alone;
+Goal 7 allows that only if a title is not content.
 
 Blocks 73: the announcement waits for this answer, since the answer may change what the package
 writes for a titled image.