diff --git a/docs/specs/2026-09-26-m5b-hardening.md b/docs/specs/2026-09-26-m5b-hardening.md
index f5e9435..4b92088 100644
--- a/docs/specs/2026-09-26-m5b-hardening.md
+++ b/docs/specs/2026-09-26-m5b-hardening.md
@@ -260,7 +260,7 @@ messages and read each post's `metadata` back through the API):
 
 | Posted | Link preview (`embeds`) | Image fetched by the server (`images`) |
 |---|---|---|
-| a plain URL | yes: a `link` embed, fetched when a client first shows it | |
+| a plain URL | marked for a preview (a `link` embed); no card appeared, because `example.com` has no OpenGraph tags, so the fetch itself was not observed | |
 | a URL in a code span | none | |
 | a URL in a fenced block | none | |
 | a Markdown link, `[text](url)` | none: only plain URLs are previewed | |
@@ -269,7 +269,8 @@ messages and read each post's `metadata` back through the API):
 | a defanged `hxxps://` URL | none | |
 
 So a code span stops both automatic fetches, and an image is fetched by the server itself, not only
-by each client: escaping images matters in every session.
+by each client: escaping images matters in every session. The owner's screenshot of the posts
+matched the readings; the test posts were then deleted.
 
 `TurnDone` gains `untrusted: bool` (a new field, so its fixture changes). Trusting `loopd`'s own
 labels here is sound: this is a defence in depth on presentation, not a trust gate.