Skip to content

Agent Callbacks

For greater control when running your agent, any-agent includes support for custom Callbacks that will be called at different points of the AnyAgent.run:

# pseudocode of an Agent run

history = [system_prompt, user_prompt]
context = Context()

while True:

    for callback in agent.config.callbacks:
        context = callback.before_llm_call(context)

    response = CALL_LLM(history)

    for callback in agent.config.callbacks:
        context = callback.after_llm_call(context)

    history.append(response)

    if response.tool_executions:
        for tool_execution in tool_executions:

            for callback in agent.config.callbacks:
                context = callback.before_tool_execution(context)

            tool_response = EXECUTE_TOOL(tool_execution)

            for callback in agent.config.callbacks:
                context = callback.after_tool_execution(context)

            history.append(tool_response)

    else:
        return response

Advanced designs such as safety guardrails or custom side-effects can be integrated into your agentic system using this functionality.

During the agent run ( agent.run_async or agent.run ), a unique Context object is created and shared across all callbacks.

any-agent populates the Context.current_span property so that callbacks can access information in a framework-agnostic way. You can check the attributes available for LLM Calls and Tool Executions in the example spans.

Implementing Callbacks

All callbacks must inherit from the base Callback class and can choose to implement any subset of the available callback methods.

You can use Context.shared to store information meant to be reused across callbacks:

from any_agent.callbacks import Callback, Context

class CountSearchWeb(Callback):
    def after_tool_execution(self, context: Context, *args, **kwargs) -> Context:
        if "search_web_count" not in context.shared:
            context.shared["search_web_count"] = 0
        if context.current_span.attributes["gen_ai.tool.name"] == "search_web":
            context.shared["search_web_count"] += 1

Callbacks can raise exceptions to stop agent execution. This is useful for implementing safety guardrails or validation logic:

class LimitSearchWeb(Callback):
    def __init__(self, max_calls: int):
        self.max_calls = max_calls

    def before_tool_execution(self, context: Context, *args, **kwargs) -> Context:
        if context.shared["search_web_count"] > self.max_calls:
            raise RuntimeError("Reached limit of `search_web` calls.")

Warning

Raising exceptions in callbacks will terminate the agent run immediately. Use this feature carefully to implement safety measures or validation logic.

Default Callbacks

any-agent comes with a set of default callbacks that will be used by default (if you don't pass a value to AgentConfig.callbacks):

If you want to disable these default callbacks, you can pass an empty list:

from any_agent import AgentConfig, AnyAgent
from any_agent.tools import search_web, visit_webpage

agent = AnyAgent.create(
    "tinyagent",
    AgentConfig(
        model_id="mistral/mistral-small-latest",
        instructions="Use the tools to find an answer",
        tools=[search_web, visit_webpage],
        callbacks=[]
    ),
)

Providing your own Callbacks

Callbacks are provided to the agent using the AgentConfig.callbacks property.

You can use get_default_callbacks:

from any_agent import AgentConfig, AnyAgent
from any_agent.callbacks import get_default_callbacks
from any_agent.tools import search_web, visit_webpage

agent = AnyAgent.create(
    "tinyagent",
    AgentConfig(
        model_id="gpt-4.1-nano",
        instructions="Use the tools to find an answer",
        tools=[search_web, visit_webpage],
        callbacks=[
            CountSearchWeb(),
            LimitSearchWeb(max_calls=3)
        ] + get_default_callbacks()
    ),
)

from any_agent import AgentConfig, AnyAgent
from any_agent.tools import search_web, visit_webpage

agent = AnyAgent.create(
    "tinyagent",
    AgentConfig(
        model_id="gpt-4.1-nano",
        instructions="Use the tools to find an answer",
        tools=[search_web, visit_webpage],
        callbacks=[
            CountSearchWeb(),
            LimitSearchWeb(max_calls=3)
        ]
    ),
)

Warning

Callbacks will be called in the order that they are added, so it is important to pay attention to the order in which you set the callback configuration.

In the above example, passing:

    callbacks=[
        LimitSearchWeb(max_calls=3)
        CountSearchWeb()
    ]

Would fail because context.shared["search_web_count"] was not set yet.

Example: Offloading sensitive information

Some inputs and/or outputs in your traces might contain sensitive information that you don't want to be exposed in the traces.

You can use callbacks to offload the sensitive information to an external location and replace the span attributes with a reference to that location:

import json
from pathlib import Path

from any_agent.callbacks.base import Callback
from any_agent.callbacks.context import Context


class SensitiveDataOffloader(Callback):

    def __init__(self, output_dir: str) -> None:
        self.output_dir = Path(output_dir)
        self.output_dir.mkdir(exist_ok=True, parents=True)

    def before_llm_call(self, context: Context, *args, **kwargs) -> Context:

        span = context.current_span

        if input_messages := span.attributes.get("gen_ai.input.messages"):
            output_file = self.output_dir / f"{span.get_span_context().trace_id}.txt"
            output_file.write_text(str(input_messages))

            span.set_attribute("gen_ai.input.messages", json.dumps(
                {"ref": str(output_file)}
            ))

        return context

You can find a working example in the Callbacks Cookbook.