AICanary

Behavioral testing for AI. Create canaries that continuously probe your models and agents for regressions, hallucinations, and policy violations.

Quickstart

Create a canary, then generate test cases from your production prompts:

bash

# 1. Create a canary
curl -X POST https://api.luxkern.com/v1/aicanary/canaries \
  -H "Authorization: Bearer lxk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{ "name": "support-bot-v2", "model": "gpt-4o" }'
 
# 2. Generate tests from a prompt
curl -X POST https://api.luxkern.com/v1/aicanary/canaries/:id/generate \
  -H "Authorization: Bearer lxk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{ "prompt": "You are a helpful customer support agent..." }'

What you get

Auto-generated behavioral test suites
Scheduled canary runs on every deploy or cron
Regression alerts when model behavior drifts
Side-by-side model comparison reports

Open AICanary Dashboard · Full API Reference