Lesson 2 Preview
A useful personal AI assistant must do more than reply to messages. It needs to understand whether the user wants information, a recommendation, a reminder, a plan or another action.
一個實用的個人 AI 助手不只需要回答訊息，還需要判斷使用者想取得資訊、獲得推薦、建立提醒、制定計劃，還是進行其他操作。
How can an AI assistant understand what the user wants it to do?
Course Journey
Lesson 2 turns safe conversation into a first decision: what does the user want next?
Lesson Story
Byte can now hold a safe conversation, but it still needs to understand what the user wants.
Can you suggest a game for me?
This is a request for a recommendation.
Remind me to bring my science book tomorrow.
This is a request to create a reminder.
Before a personal AI assistant chooses an action, it first needs to identify the user's intent.
Byte 現在已經能進行安全對話，但它仍然需要理解使用者真正想做甚麼。在個人 AI 助手選擇下一步行動之前，它需要先判斷使用者的意圖。
What the user wants the assistant to do.
Choosing the best label for a message.
A short name for one kind of request.
A clear format a program can use.
A short explanation for the choice.
What the assistant should do after classifying.
Frontend Activity
Try classifying the message yourself first. This activity runs only in the browser and does not send answers to the server.
Message
Choose one intent
Live AI Preview
Choose an assistant context, add one short rule, and ask Byte to return an intent, reason, and suggested next action.
Safety first: do not enter your real name, password, address, phone number or other private information.
安全第一：請勿輸入真實姓名、密碼、地址、電話號碼或其他私人資料。
Step 1
The browser can choose only from these server-controlled contexts.
Step 2
Add one small rule without replacing the full classifier prompt.
180 characters left
Step 3
Returns one allowed intent and a suggested next step.
Keep it under 300 characters and avoid private information.
Byte is classifying...
Suggested test messages
Detected Intent
Prompt Design
Classify this message.
The AI may answer in many different formats.
Choose exactly one allowed intent, give one short reason, and suggest one next action.
The program can use the result more reliably.
Did the AI choose the same intent as you?
The public demo does not save classifications or conversations. It only returns a short result for the message you submit.
公開 Demo 不會儲存分類結果或對話，只會為你提交的訊息回傳簡短結果。
Routing
Lesson 2 identifies what the user wants. Lesson 3 will connect each intent to a different assistant skill.
This demo does not create real reminders, save preferences, create calendar events, or store user information.
Student Project
Students customize assistant context, classification rules, test messages, intent explanations, and suggested next actions.
Why This Matters
A personal AI assistant needs structure before it can act. Intent classification lets a program decide whether to answer, recommend, plan, remember a preference, or ask a follow-up question.
個人 AI 助手需要先理解意圖，程式才可以決定要回答、推薦、制定計劃、記住喜好，還是追問更多資料。
Lesson 3 Preview
Understanding intent is only the beginning. In Lesson 3, students connect each intent to a different assistant skill, such as answering questions, making recommendations or creating simple plans.