Files
zooqueenandHanzo Dev b00044ece2 Check each recipe's answer against the obvious one
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]>
2026-10-09 17:39:13 -07:00
..

Customer Support: Churn Risk

Ask Kai whether a customer's support message says they may leave, and whether it needs a leader's attention.

Request (request.json)

{
  "model": "kai",
  "state": {
    "customer_mrr": 450,
    "tenure_months": 14,
    "message": "We have been dealing with API 502 errors all week. We cannot afford this downtime and are looking into moving our stack to AWS Bedrock if this isn't resolved today."
  },
  "questions": {
    "churn_propensity": {
      "type": "noul",
      "instructions": "Is this customer actively planning to cancel or switch providers?"
    },
    "executive_escalation": {
      "type": "choice",
      "instructions": "Should an engineering leader or VP jump on this ticket?",
      "criteria": {
        "standard_support": "standard Tier-2 support resolution",
        "executive_outreach": "immediate personal email/call from Founder or Head of Customer Success"
      }
    }
  }
}

Checked answers (expect.json)

  • churn_propensity (noul): P(true) above 0.5
  • executive_escalation (choice): not checked, no obvious answer on this input

Run

From the repository root, after pip install -r requirements.txt and npm install:

curl -sS --fail-with-body https://api.hanzo.ai/v1/decisions \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d @recipes/07_support_churn_risk/request.json

python3 recipes/07_support_churn_risk/run.py
npx tsx recipes/07_support_churn_risk/run.ts

run.py and run.ts print every answer and exit 1 when a checked answer is wrong.