From 898be465a38e402ee3547b25e9172417656540bf Mon Sep 17 00:00:00 2001
From: teamchong <25894545+teamchong@users.noreply.github.com>
Date: Tue, 9 Jun 2026 21:45:12 -0400
Subject: [PATCH] docs: README example image now matches real dashboard density
 (full dense page, synthetic content)

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 676d135..0323d97 100644
--- a/README.md
+++ b/README.md
@@ -14,7 +14,7 @@ actual). Single sessions measure ~68%.
 
 This is what the model sees instead of text:
 
-![example: 5,000 chars of TypeScript rendered as one 908×880 PNG](docs/assets/example-render.png)
+![example: ~17k chars of tool-result JSON + prose rendered as one dense 1573×1280 page (what the dashboard shows)](docs/assets/example-render.png)
 
 *5,000 characters of source code — ≈1,250 text tokens as a transcript, ≈530
 image tokens as this PNG. The model reads it at 100/100 on a clean eval (see