eval-tool-use

What problem does it solve?

Evaluate tool-use quality in LLM agents by assessing tool selection and argument construction. This guidance helps separate deterministic tool interactions from higher-level reasoning and promotes robust evaluation workflows.

Core Features & Use Cases

• Tool selection audits: verify the agent chooses the correct tool for a given task.
• Argument validation: ensure inputs to tools conform to expected schemas and constraints.
• Sequencing and error handling: verify the order of tool calls and recovery after failures.
• Use Case: Test a Claude agent with multiple tools to confirm it calls the right tool with appropriate arguments and handles tool errors gracefully.

Quick Start

Provide a trace of a tool-use session and run the evaluation to identify tool selection, argument, and sequencing issues.