diff --git a/PLAN.md b/PLAN.md
index 87e68fa..1a89fba 100644
--- a/PLAN.md
+++ b/PLAN.md
@@ -653,8 +653,9 @@ would justify starting it, and none starts without an explicit launch command.
 - **A run id in the ML worker protocol.** Condition: the day two requests can
   be in flight at once (for example, an analysis that runs during training).
   Until then V47's fence is enough, and the limits page says so.
-- **Widen the text vocabulary.** V24 capped it; V25 measured that the memory
-  guard made the typed-array rewrite unnecessary. Condition: a real dataset
+- **Widen the text vocabulary.** V24 capped it; V25 measured that the pipeline
+  was never the bottleneck, so the typed-array rewrite was dropped and a named
+  memory guard was added instead. Condition: a real dataset
   where the cap visibly costs accuracy, measured before and after.
 - **V12, consented generative chat.** Postponed by the owner (09/10/2026).
   Condition: a product decision, plus a fresh read of the provider's data
diff --git a/README.md b/README.md
index 562b2a8..db5cd83 100644
--- a/README.md
+++ b/README.md
@@ -341,8 +341,9 @@ CI builds, tests and deploys on every push: pull requests get a Cloudflare Pages
 
 ## Roadmap
 
-Development proceeds in planned "caps" of feature waves; seven caps have shipped (MVP
-through the lab meeting the real world: real photos, real text, real file sizes). The last, Cap 7,
+Development proceeds in planned "caps" of feature waves; seven caps have shipped. The
+first six took it from the MVP to the lab meeting the real world (real photos, real text, real
+file sizes); the last, Cap 7,
 made what already ships hold up under pressure: worker lifecycle, damaged share links,
 CI hardening, metric edge cases, a store refactor, and major upgrades taken one at a time
 (Vitest 5 and vite-plugin-pwa 2 shipped; TypeScript 7 typechecks this code with no error