Skip to content

Logger

Structured logging with automatic session context inclusion for tools and orchestrators.

Logger Class

::: agenticai_core.runtime.sessions.request_context.Logger options: show_root_heading: true show_source: false members_order: source

Basic Usage

from agenticai_core.runtime.sessions.request_context import Logger

@Tool.register(name="my_tool", description="Tool with logging")
async def my_tool(operation: str):
    logger = Logger('MyTool')

    await logger.info(f"Starting operation: {operation}")

    try:
        # Your tool logic here
        result = {"success": True, "data": "processed"}
        await logger.info("Operation completed successfully")
        return result

    except Exception as e:
        await logger.error(f"Operation failed: {str(e)}")
        raise

With RequestContext

from agenticai_core.runtime.sessions.request_context import RequestContext, Logger

@Tool.register(name="integrated_tool", description="Tool with shared context")
async def integrated_tool():
    # Create shared context
    ctx = RequestContext()
    logger = Logger('IntegratedTool', request_context=ctx)

    # Access environment and memory through same context
    env = await ctx.get_environment_variables()
    memory = ctx.get_memory()

    await logger.info("Tool execution completed")
    return {"success": True}

Log Levels

logger = Logger('MyTool')

await logger.debug("Detailed diagnostic information")
await logger.info("General operational information")  
await logger.warning("Potential issue requiring attention")
await logger.error("Error affecting functionality")

Best Practices

  1. Use descriptive logger names: Logger('UserAuthTool') instead of Logger('Tool')
  2. Include context: f"Processing payment: amount=${amount}, user={user_id}"
  3. Don't log sensitive data: Sanitize or redact passwords, API keys, tokens
  4. Use appropriate levels: Debug for diagnostics, info for operations, warning/error for issues

Troubleshooting

Common Issues

  1. Logger Not Capturing Session Context: 

    # ✅ Ensure Logger is created within request context
ctx = RequestContext()
logger = Logger('MyTool', request_context=ctx)

  2. Missing Log Output: 

    # Check if async/await is used correctly
await logger.info("Message")  # ✅ Correct
logger.info("Message")        # ❌ Missing await

  3. Performance Issues with Logging: 

    # Avoid expensive operations in log messages
# Use conditional logging for debug messages
if debug_enabled:
    await logger.debug(expensive_debug_info())

Related resources