feat: chat uploads, vision turn isolation, and MindSpace agent improvements

Add chat file/image upload UX, attachment proxying, vision thumbnails, and per-turn image scoping so agents only use the current upload. Extend MindSpace asset context, billing token state, OA/scenario verify scripts, and related runtime config. Co-authored-by: Cursor <cursoragent@cursor.com>
2026-07-11 00:23:01 +08:00
parent e3063ea806
commit 32fb2cdeaf
51 changed files with 4588 additions and 186 deletions
@@ -0,0 +1,28 @@
+{
+  "id": "john4-children-hobby-diet-update",
+  "name": "john4 儿童爱好问卷增加饮食偏好",
+  "description": "在已有 page-data-collect 问卷上追加 q4_diet_meals（每日几顿），验证 Agent 交付 + Page Data 可提交",
+  "account": {
+    "username": "john4",
+    "password": "888888"
+  },
+  "steps": [
+    {
+      "action": "login",
+      "label": "登录 john4"
+    },
+    {
+      "action": "chat",
+      "label": "问卷增加饮食偏好（每日几顿）",
+      "message": "问卷页面再增加一个饮食偏好，比如每日几顿这种",
+      "selectedChatSkill": "page-data-collect",
+      "expect": {
+        "assistantMinChars": 80,
+        "timeoutMs": 600000,
+        "replyKeywords": ["问卷", "饮食"],
+        "forbidReplyPatterns": ["8899", "127.0.0.1:", "PLACEHOLDER_PAGE_ID", "survey-api"],
+        "surveyDietUpdate": true
+      }
+    }
+  ]
+}