From 7e885faf83ed7cfdefd996b2c897fe1852945564 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=85=AC=E6=98=8E?=
 <83812544+Ed1s0nZ@users.noreply.github.com>
Date: Sun, 28 Dec 2025 15:19:23 +0800
Subject: [PATCH] Update README.md

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

diff --git a/README.md b/README.md
index 511dd185..cb9d59fd 100644
--- a/README.md
+++ b/README.md
@@ -8,6 +8,7 @@
 [中文](README_CN.md) | [English](README.md)
 
 CyberStrikeAI is an **AI-native penetration-testing copilot** built in Go. It combines hundreds of security tools, MCP-native orchestration, and an agent that reasons over findings so that a full engagement can be run from a single conversation.
+> In security, what is truly scarce is not tools, but judgment; judgment is born from experience, and experience is often bound to individuals, difficult to inherit and difficult to reuse. CyberStrikeAI does not attempt to automate attacks, but instead focuses on a harder problem: in complex, dynamic, and uncertain environments, what should be done next, and why. It does not pursue more aggressive automation; rather, it seeks to transform the way security experts think—their decision paths and lessons learned from failure—into a system capability that is constrained, auditable, and evolvable. If experience can exist beyond individuals, then security can finally become something that can be inherited by systems.
 
 
 ## Interface & Integration Preview