From c76bd715db6f57d54fa126589b862eb34be7a58c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Stig-=C3=98rjan=20Smelror?= <smelror@gmail.com>
Date: Sat, 28 Feb 2026 23:22:07 +0100
Subject: [PATCH] docs(home): highlight benchmark results on landing page

Replace hero copy with 250-turn benchmark findings and context-cost impact.
Update secondary CTA to link directly to the benchmark post.
---
 src/pages/index.astro | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/pages/index.astro b/src/pages/index.astro
index f25c6bc..26a027c 100644
--- a/src/pages/index.astro
+++ b/src/pages/index.astro
@@ -30,14 +30,14 @@ const version = pkg.version ?? "";
             <span class="meta-item">NO CLOUD REQUIRED</span>
           </div>
           <p class="hero-description">
-            Zaguán Blade + Zaguán Coder Daemon are two pieces of one whole. Running Blade through zcoderd unlocks its full potential - longer system prompts, more context, better results. Local AI works, but it's limited.
+            Zaguán Blade + zcoderd are built for long-running coding sessions. In our 250-turn benchmark, context compression and selective retrieval reduced token cost by ~31% on average, with higher gains in context-heavy runs. It’s not always cheaper in lean sessions — it’s designed to prevent context bloat when history grows.
           </p>
           <div class="hero-actions">
             <a href="#download" class="action-primary">
               <span class="action-text">RUN AI LOCALLY. START HERE.</span>
               <span class="action-arrow">→</span>
             </a>
-            <a href="#philosophy" class="action-secondary">WHY WE BUILT IT</a>
+            <a href="/blog/measuring-token-savings" class="action-secondary">SEE THE BENCHMARK</a>
           </div>
         </div>
       </div>