Back to Prompt Library
testing

Integrate Giskard for Evaluation

Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.

Linked challenge: Build an AI-Powered Regulatory Compliance Risk Agent

Format
Text-first
Lines
1
Sections
1
Linked challenge
Build an AI-Powered Regulatory Compliance Risk Agent

Prompt source

Original prompt text with formatting preserved for inspection.

1 lines
1 sections
No variables
0 checklist items
Integrate Giskard into your agent's development loop. Specifically, create a Giskard test suite to evaluate the 'policy_evaluation_tool's accuracy. For instance, provide scenarios where a specific action (e.g., 'transfer data to non-EU server without SCCs') should 'FAIL' under GDPR, and ensure your agent's tool correctly identifies this. Report the Giskard evaluation results.

Adaptation plan

Keep the source stable, then change the prompt in a predictable order so the next run is easier to evaluate.

Keep stable

Preserve the rubric, target behavior, and pass-fail criteria as the baseline for evaluation.

Tune next

Adjust fixtures, mocks, and thresholds to the system under test instead of weakening the assertions.

Verify after

Make sure the prompt catches regressions instead of just mirroring the happy-path examples.