diff --git a/docs/e2e-skills-benchmark.md b/docs/e2e-skills-benchmark.md
index 674a3b4b..c82ca61b 100644
--- a/docs/e2e-skills-benchmark.md
+++ b/docs/e2e-skills-benchmark.md
@@ -7,7 +7,7 @@ This benchmark validates the meta skill workflow for capability-gap discovery, C
 - Domain: skill discovery + installation + update
 - Focus: `skills/meta-skill-installer`
 - Providers: default `kimi-coding` (override with `PROVIDERS`)
-- Cases: 4
+- Cases: 5
 
 Case prompts are stored in:
 - `scripts/e2e-skills-benchmark/cases/`
@@ -20,6 +20,7 @@ The case set references real public pages from ClawHub:
 - [Home Assistant](https://clawhub.ai/skills/homeassistant)
 - [CodexMonitor](https://clawhub.ai/odrobnik/codexmonitor)
 - [Spotify (gap-discovery UX flow)](https://clawhub.ai/search?q=spotify)
+- [Notion (gap-discovery UX flow)](https://clawhub.ai/search?q=notion)
 
 ## Prerequisites
 
@@ -87,7 +88,8 @@ For each run:
    - `clawhub install`
    - `review-skill-security.mjs`
    - for case 03 also `clawhub update`
-   - for case 04, final response must include ClawHub + install confirmation language, and must not run `clawhub install/update` before confirmation
+   - for case 04, prompt is a natural user request only; agent must self-discover capability gap, propose ClawHub + security review + install confirmation, and must not run workaround commands (`osascript`, `ha.sh`, `spogo`, `spotify_player`) before user confirmation
+   - for case 05, prompt is a natural Notion request; agent must discover missing capability, search skill candidates, trigger `install_guard` (blocked until confirmation), and ask for explicit install consent plus token/auth prerequisites
 
 ## Notes
 
diff --git a/scripts/e2e-skills-benchmark/analyze.mjs b/scripts/e2e-skills-benchmark/analyze.mjs
index 0eaee0ed..3f621328 100755
--- a/scripts/e2e-skills-benchmark/analyze.mjs
+++ b/scripts/e2e-skills-benchmark/analyze.mjs
@@ -66,6 +66,7 @@ const CASE_RULES = {
   "case-04-gap-discovery-spotify-ux": {
     requireExecUsage: false,
     requiredResponseRegex: [
+      "缺少|没有.*(技能|能力|集成)|capability gap",
       "clawhub|cloud\\s*hub|cloudhub",
       "安装|install",
       "是否|要不要|would you like|do you want",
@@ -74,6 +75,34 @@ const CASE_RULES = {
     forbiddenCommandTokens: [
       ["clawhub", "install"],
       ["clawhub", "update"],
+      ["osascript"],
+      ["spogo"],
+      ["spotify_player"],
+      ["ha.sh"],
+      ["/api/states"],
+    ],
+  },
+  "case-05-gap-discovery-notion-ux": {
+    requireExecUsage: false,
+    requiredCommandTokens: [
+      ["clawhub", "search"],
+      ["notion"],
+    ],
+    requiredEventTokens: [
+      ["install_guard", "blocked"],
+    ],
+    requiredResponseRegex: [
+      "notion",
+      "安装|install",
+      "是否|要不要|would you like|do you want|同意",
+      "token|授权|integration",
+    ],
+    forbiddenCommandTokens: [
+      ["osascript"],
+      ["spogo"],
+      ["spotify_player"],
+      ["ha.sh"],
+      ["/api/states"],
     ],
   },
 };
@@ -324,6 +353,22 @@ for (let i = 1; i < rows.length; i++) {
       }
     }
 
+    if (Array.isArray(rules.requiredEventTokens)) {
+      const eventLines = events.map((event) => JSON.stringify(event).toLowerCase());
+      for (let r = 0; r < rules.requiredEventTokens.length; r++) {
+        const tokenList = rules.requiredEventTokens[r];
+        const passed = eventLines.some((line) =>
+          tokenList.every((token) => line.includes(token.toLowerCase())),
+        );
+        addCheck(
+          analysis,
+          `event-${r + 1}`,
+          `event log contains tokens: ${tokenList.join(" + ")}`,
+          passed,
+        );
+      }
+    }
+
     if (Array.isArray(rules.forbiddenCommandTokens)) {
       for (let r = 0; r < rules.forbiddenCommandTokens.length; r++) {
         const tokenList = rules.forbiddenCommandTokens[r];
diff --git a/scripts/e2e-skills-benchmark/cases/case-04-gap-discovery-spotify-ux.txt b/scripts/e2e-skills-benchmark/cases/case-04-gap-discovery-spotify-ux.txt
index 5c8a7c16..bf2b917a 100644
--- a/scripts/e2e-skills-benchmark/cases/case-04-gap-discovery-spotify-ux.txt
+++ b/scripts/e2e-skills-benchmark/cases/case-04-gap-discovery-spotify-ux.txt
@@ -1,10 +1 @@
-请模拟真实用户的首轮请求：
-
-用户说：「随机播放 spotify 中的音乐」。
-
-要求：
-1. 先明确你当前缺少直接控制 Spotify 的能力（能力缺口识别）。
-2. 不要直接拒绝结束，要给出恢复路径：引导用户可通过 ClawHub 搜索并安装相关 skill。
-3. 明确说明会先做安全审查，再安装。
-4. 在同一轮里必须询问用户是否要继续安装（显式确认），在用户确认前不要执行 `clawhub install` 或 `clawhub update`。
-5. 输出用中文，且要包含关键词：ClawHub、安全审查、安装、是否要继续。
+随机播放 spotify 中的音乐
diff --git a/scripts/e2e-skills-benchmark/cases/case-05-gap-discovery-notion-ux.txt b/scripts/e2e-skills-benchmark/cases/case-05-gap-discovery-notion-ux.txt
new file mode 100644
index 00000000..aff66735
--- /dev/null
+++ b/scripts/e2e-skills-benchmark/cases/case-05-gap-discovery-notion-ux.txt
@@ -0,0 +1 @@
+帮我在 Notion 新建一个页面，标题是今天待办，并写入三条任务：修复登录 bug、写周报、安排评审