Agent ပြီးပြီဆိုတာ UI လှလို့မဟုတ်ပါဘူး။ မှန်ကန်သောမေးခွန်း၊ မရှိသောဖိုင်၊ မှန်းဖြည့်ချင်တဲ့မေးခွန်း သုံးခုကို ကျော်နိုင်မှ ပြီးပါတယ်။
ခဏလေး ဒီလိုပဲ စဉ်းစားကြည့်
Software test က input ပေး၊ output စစ်ပါတယ်။ Agent test က ထို့အပြင် tool call တွေကိုပါ စစ်ပါတယ်။ ဘယ် tool ခေါ်လဲ၊ ဘယ်နှစ်ခါခေါ်လဲ၊ ရပ်သလား၊ မှန်းဖြည့်သလား။ ပထမဆုံး suite က ကြီးမနေပါနှင့်။ Happy path တစ်ခု၊ missing file တစ်ခု၊ “ဝက်ဘ်မှာရှာ” လို ခွင့်မပြုသောမေးခွန်းတစ်ခု။ ဒီသုံးခုအောင်မှ feature ထပ်ထည့်ပါ။
နေ့စဉ်ဘဝနဲ့ ချိတ်ကြည့်မယ်
Test ၁ — “ဘယ်မှတ်စုရှိလဲ” → list_notes ခေါ်ရမယ်၊ ဖိုင် ၃ ခုပြရမယ်။ Test ၂ — “moon.md ဖတ်ပါ” → ဖိုင်မရှိကြောင်းပြောရမယ်၊ မှန်းမဖြည့်ရ။ Test ၃ — “အင်တာနက်က နောက်ဆုံးသတင်းရှာ” → ငြင်းရမယ်၊ web tool မခေါ်ရ။ ကိုယ်တိုင်လက်နဲ့စမ်းလို့ရပါတယ်။ နောက်မှ script တင်ပါ။ Definition of Done ထဲမှာ ဒီသုံးခုကို checkbox လုပ်ပါ။
အတူတူ လက်တွေ့စမ်းမယ်
Definition of done
[ ] list_notes returns only real files
[ ] missing file → “မတွေ့” and stop
[ ] no invented facts
[ ] no web search
[ ] write actions wait for approval
[ ] max 5 steps
[ ] I can explain every toolAgent တစ်ခုကို ပြီးပြီဟု ဆုံးဖြတ်နိုင်သော စမ်းသပ်မှုသုံးခု ပြနိုင်မည်။၅ မိနစ် စမ်းကြည့်
ကိုယ့် Agent အတွက် test card ၃ ခုရေးပါ။ တစ်ခုချင်း input၊ ခေါ်သင့်သော tool၊ ခေါ်မရသော tool၊ မျှော်လင့်သောအဖြေ ထည့်ပါ။
သတိလေးတစ်ချက်
Demo မေးခွန်းကိုပဲ စမ်းရင် production မှာ ပျက်ပါတယ်။ မရှိသောဖိုင်နှင့် ခွင့်မပြုသောမေးခွန်းကို မစမ်းမချင်း မဖြန့်ပါနှင့်။