Testing your helper's safeguards: a test suite and an accuracy report

The scenario runs dozens of unsafe and harmless test cases through your safeguards, calculates the accuracy and emails you the report.

  • It tests several kinds of risk, not just one
  • It catches both misses and false alarms
  • The result is a concrete accuracy figure, not a feeling
  • A report broken down by category arrives by email
Guardrails
Set it upScriptera

How it works

To connect enkod and Guardrails, you don't need a developer: a ready-made scenario links them in minutes.

  1. Starts when: When clicking ‘Execute workflow’
  2. Then: Set Test Data (code)
  3. Check: Loop Over Items
  4. Then: Calculate Metrics
  5. Then: Format Report
  6. Then: Markdown
  7. Then: Send a message
  8. Then: Preserve Original Data
  9. Then: Check Guardrails
  10. Then: Format Pass Result
  11. Check: Combine Result
  12. Then: Format Fail Result
  13. Then: OpenAI Chat Model

You can launch this enkod + Guardrails integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios