mirror of
https://github.com/hanzoai/kai.git
synced 2026-11-03 01:05:42 +00:00
zooqueenandHanzo Dev
b00044ece2
Every runner printed whatever came back and exited 0, and fell back to a second raw HTTP call when the SDK import failed. Now each recipe carries expect.json, the answers a careful reader gives on its own input: a list of acceptable options for a choice, of level indices for a score, true or false for a noul. run.py and run.ts send request.json through the SDK, print every answer, and exit 1 when a checked one is wrong. A question with no obvious answer on its input is printed and not checked. Recipes 09 and 11 had no input with an obvious answer, so each state gains the one fact that decides it. The recipe READMEs describe what each recipe asks instead of capabilities it never exercised, and copy the request and checked answers from the two JSON files. Co-authored-by: Hanzo Dev <[email protected]>
24 lines
782 B
JSON
24 lines
782 B
JSON
{
|
|
"model": "kai",
|
|
"state": {
|
|
"prospect": "VP Engineering at MedTech Co",
|
|
"history": [
|
|
"downloaded whitepaper on HIPAA compliance with private models",
|
|
"attended webinar on agent security",
|
|
"created free developer account and tested 50 API queries",
|
|
"asked in the webinar Q&A for a live walkthrough with their security team"
|
|
]
|
|
},
|
|
"questions": {
|
|
"next_action": {
|
|
"type": "choice",
|
|
"instructions": "What is the optimal next engagement step?",
|
|
"criteria": {
|
|
"book_demo": "direct calendar link for technical demo with Solutions Architect",
|
|
"send_case_study": "email healthcare case study and security audit results",
|
|
"invite_slack": "invite to private developer Slack channel"
|
|
}
|
|
}
|
|
}
|
|
}
|