Logging tool calls and results in Agentic Ai

Introduction

Logging tool calls and results helps you keep track of what your tools do and what they return. This makes it easier to find problems and understand how your system works.

When you want to see what commands your tools are running step-by-step.

When you need to check if a tool worked correctly or failed.

When you want to save a history of tool actions for later review.

When debugging a problem to find where it happened.

When monitoring system behavior during automated tasks.

Syntax

Agentic_ai

log_tool_call(tool_name, input_parameters)
log_tool_result(tool_name, result_data)

log_tool_call records the tool name and what inputs it received.

log_tool_result records the output or result from the tool.

Examples

This logs that the BackupTool was called with the path '/data' and returned 'Success'.

Agentic_ai

log_tool_call("BackupTool", {"path": "/data"})
log_tool_result("BackupTool", "Success")

This logs a deployment attempt with version 1.2.3 and a failure with a timeout error.

Agentic_ai

log_tool_call("DeployTool", {"version": "1.2.3"})
log_tool_result("DeployTool", {"status": "failed", "error": "Timeout"})

Sample Program

This example shows logging a tool call and its result for a tool named TestTool.

Agentic_ai

log_tool_call("TestTool", {"input": "test data"})
log_tool_result("TestTool", "Output OK")

Important Notes

Always log both the call and the result to have a complete picture.

Use clear and consistent names for tools to avoid confusion.

Store logs in a safe place where you can easily search and review them.

Summary

Logging tool calls and results helps track what tools do and their outputs.

This practice is useful for debugging, monitoring, and auditing.

Use simple functions to log inputs and outputs clearly and consistently.