AG-001Agent · Understanding the joblayaeval.IntentResolution
Intent Resolution
What does the user want — info, an action, or a fix?
Simulation state:
When it passes
Sorted quickly and surely.
Threshold satisfied: evaluation score >= 0.70.
When it fails
Unsure, lands in the wrong bucket.
Violation detected: execution flagged for human review or fallback routing.
SDK Specification
| Parameter | Value | Description |
|---|---|---|
| Class | IntentResolution | Python evaluator class in layaeval |
| Required Fields | "query" | Arguments expected in the call signature |
| Default Threshold | >= 0.70 | Passes when quality score is at or above this value |
| Category | Agent Evaluation | Understanding the job (Did the agent get what you wanted, and finish it?) |
Python Example
intent-resolution.py
1from layaeval import IntentResolution23metric = IntentResolution(threshold=0.70) # passes at or above 0.7045res = metric(6 query="What is the capital of France?",7)89print(res.passed, res.value, res.confidence)
Related Agent Checks