AG-007Agent · Using toolslayaeval.ToolCallAccuracy

Tool Call Accuracy

Was the tool called the right way?

Simulation state:

When it passes

Clean, correct call.

Threshold satisfied: evaluation score >= 3.0.

When it fails

A broken field.

Violation detected: execution flagged for human review or fallback routing.

SDK Specification

Parameter	Value	Description
Class	ToolCallAccuracy	Python evaluator class in layaeval
Required Fields	"task", "tool_definition", "tool_call"	Arguments expected in the call signature
Default Threshold	>= 3.0	Passes when quality score is at or above this value
Category	Agent Evaluation	Using tools (Right tool, right inputs, results actually used.)

Python Example

pip install layaeval

tool-call-accuracy.py

1from layaeval import ToolCallAccuracy
2
3metric = ToolCallAccuracy(threshold=3.0)  # passes at or above 3.0
4
5res = metric(
6    task="Book a hotel in Paris for 2 guests on Mar 4.",
7    tool_definition="search_hotels(city: str, date: str, guests: int)",
8    tool_call="search_hotels(city='Paris', date='2026-03-04', guests=2)",
9)
10
11print(res.passed, res.value, res.confidence)