Events
Adapted from the Griptape AI Framework documentation.
Overview
You can configure the global EventBus with EventListeners to listen for various framework events. See Event Listener Drivers for examples on forwarding events to external services.
Specific Event Types
You can listen to specific event types:
from griptape.events import ( BaseEvent, EventBus, EventListener, FinishActionsSubtaskEvent, FinishPromptEvent, FinishTaskEvent, StartActionsSubtaskEvent, StartPromptEvent, StartTaskEvent, ) from griptape.structures import Agent def on_event(event: BaseEvent) -> None: print(event.__class__) EventBus.add_event_listeners( [ EventListener( on_event, event_types=[ StartTaskEvent, FinishTaskEvent, StartActionsSubtaskEvent, FinishActionsSubtaskEvent, StartPromptEvent, FinishPromptEvent, ], ) ] ) agent = Agent() agent.run("tell me about griptape")
<class 'griptape.events.start_task_event.StartTaskEvent'>
[02/27/25 20:29:18] INFO PromptTask b1dc556de12242ab911b8cc7a8e448f0
Input: tell me about griptape
<class 'griptape.events.start_prompt_event.StartPromptEvent'>
<class 'griptape.events.finish_prompt_event.FinishPromptEvent'>
<class 'griptape.events.finish_task_event.FinishTaskEvent'>
[02/27/25 20:29:22] INFO PromptTask b1dc556de12242ab911b8cc7a8e448f0
Output: Grip tape is a sandpaper-like sheet that is
applied to the top surface of a skateboard deck to
provide traction between the rider's shoes and the
board. It is an essential component for
skateboarders, as it helps them maintain control
and stability while performing tricks and
maneuvers. Grip tape is typically made from a
durable, gritty material with an adhesive backing
that allows it to stick securely to the deck.
Here are some key points about grip tape:
1. **Material**: Most grip tape is made from
silicon carbide or aluminum oxide, which provides
the rough texture needed for grip.
2. **Application**: Applying grip tape involves
peeling off the backing and carefully placing it on
the skateboard deck, ensuring there are no air
bubbles. Excess tape is usually trimmed off with a
sharp blade.
3. **Customization**: Grip tape comes in various
colors, patterns, and designs, allowing
skateboarders to personalize their boards. Some
brands offer die-cut designs or printed graphics.
4. **Maintenance**: Over time, grip tape can wear
down, especially in areas of high use. It may need
to be replaced periodically to maintain optimal
grip.
5. **Other Uses**: While primarily used for
skateboards, grip tape can also be applied to other
surfaces where additional traction is needed, such
as scooter decks or tool handles.
Overall, grip tape is a crucial element for
skateboard performance, providing the necessary
friction for safe and effective riding.
All Event Types
Or listen to all events:
from griptape.events import BaseEvent, EventBus, EventListener from griptape.structures import Agent def handler1(event: BaseEvent) -> None: print("Handler 1", event.__class__) def handler2(event: BaseEvent) -> None: print("Handler 2", event.__class__) EventBus.add_event_listeners( [ EventListener(handler1), EventListener(handler2), ] ) agent = Agent() agent.run("tell me about griptape")
Handler 1 <class 'griptape.events.start_structure_run_event.StartStructureRunEvent'>
Handler 2 <class 'griptape.events.start_structure_run_event.StartStructureRunEvent'>
Handler 1 <class 'griptape.events.start_task_event.StartTaskEvent'>
Handler 2 <class 'griptape.events.start_task_event.StartTaskEvent'>
[02/27/25 20:28:19] INFO PromptTask 167acee86fc948cb953404e8ec537bb5
Input: tell me about griptape
Handler 1 <class 'griptape.events.start_prompt_event.StartPromptEvent'>
Handler 2 <class 'griptape.events.start_prompt_event.StartPromptEvent'>
Handler 1 <class 'griptape.events.finish_prompt_event.FinishPromptEvent'>
Handler 2 <class 'griptape.events.finish_prompt_event.FinishPromptEvent'>
Handler 1 <class 'griptape.events.finish_task_event.FinishTaskEvent'>
Handler 2 <class 'griptape.events.finish_task_event.FinishTaskEvent'>
[02/27/25 20:28:22] INFO PromptTask 167acee86fc948cb953404e8ec537bb5
Output: Grip tape is a gritty, sandpaper-like
material that is applied to the top surface of a
skateboard or scooter deck to provide traction for
the rider's feet. It is essential for performing
tricks and maintaining control while riding. Here
are some key points about grip tape:
1. **Material**: Grip tape is typically made from
silicon carbide or aluminum oxide, which gives it
its rough texture. The backing is usually a strong
adhesive that sticks firmly to the deck.
2. **Application**: To apply grip tape, you peel
off the backing and carefully place it on the deck,
smoothing out any air bubbles. Excess tape is
trimmed off around the edges of the deck.
3. **Variety**: Grip tape comes in various colors,
designs, and even custom prints, allowing riders to
personalize their boards. Some grip tapes also
feature perforations to prevent air bubbles during
application.
4. **Durability**: Over time, grip tape can wear
down, especially in areas of high friction. It may
need to be replaced periodically to maintain
optimal grip.
5. **Functionality**: Besides providing grip, the
rough surface of the tape helps riders perform
tricks by allowing them to "stick" to the board
better during maneuvers like ollies and flips.
6. **Maintenance**: Keeping grip tape clean can
extend its life. Dirt and debris can be removed
using a brush or grip tape cleaner.
Grip tape is a crucial component for skateboarders
and scooter riders, enhancing both performance and
safety by ensuring a secure footing.
Handler 1 <class 'griptape.events.finish_structure_run_event.FinishStructureRunEvent'>
Handler 2 <class 'griptape.events.finish_structure_run_event.FinishStructureRunEvent'>
Stream Iterator
You can use Structure.run_stream() for streaming Events from the Structure in the form of an iterator.
Tip
Set stream=True on your Prompt Driver in order to receive completion chunk events.
from griptape.events import BaseEvent from griptape.structures import Agent agent = Agent() for event in agent.run_stream("Hi!", event_types=[BaseEvent]): # All Events print(type(event))
[02/27/25 20:29:24] INFO PromptTask cc5206dcacd74046a2063b8fd8f61914
Input: Hi!
<class 'griptape.events.start_structure_run_event.StartStructureRunEvent'>
<class 'griptape.events.start_task_event.StartTaskEvent'>
<class 'griptape.events.start_prompt_event.StartPromptEvent'>
[02/27/25 20:29:28] INFO PromptTask cc5206dcacd74046a2063b8fd8f61914
Output: Hello! How can I assist you today?
<class 'griptape.events.finish_prompt_event.FinishPromptEvent'>
<class 'griptape.events.finish_task_event.FinishTaskEvent'>
Context Managers
You can also use EventListeners as a Python Context Manager.
The EventListener will automatically be added and removed from the EventBus when entering and exiting the context.
from griptape.events import EventBus, EventListener, FinishStructureRunEvent, StartPromptEvent from griptape.structures import Agent EventBus.add_event_listeners( [EventListener(lambda e: print(f"Out of context: {e.type}"), event_types=[StartPromptEvent])] ) agent = Agent(input="Hello!") with EventListener(lambda e: print(f"In context: {e.type}"), event_types=[FinishStructureRunEvent]): agent.run() agent.run()
[02/27/25 20:28:40] INFO PromptTask 0f1dcd9263ef4c6caade936c2a31c041
Input: Hello!
Out of context: StartPromptEvent
[02/27/25 20:28:41] INFO PromptTask 0f1dcd9263ef4c6caade936c2a31c041
Output: Hello! How can I assist you today?
In context: FinishStructureRunEvent
INFO PromptTask 0f1dcd9263ef4c6caade936c2a31c041
Input: Hello!
Out of context: StartPromptEvent
[02/27/25 20:28:42] INFO PromptTask 0f1dcd9263ef4c6caade936c2a31c041
Output: Hi there! How can I help you today?
Streaming
You can use the BaseChunkEvent to stream the completion results from Prompt Drivers.
from griptape.drivers.prompt.openai import OpenAiChatPromptDriver from griptape.events import BaseChunkEvent, EventBus, EventListener from griptape.structures import Pipeline from griptape.tasks import PromptTask from griptape.tools import PromptSummaryTool, WebScraperTool EventBus.add_event_listeners( [ EventListener( lambda e: print(str(e), end="", flush=True), event_types=[BaseChunkEvent], ), ] ) pipeline = Pipeline() pipeline.add_tasks( PromptTask( "Based on https://griptape.ai, tell me what griptape is.", prompt_driver=OpenAiChatPromptDriver(model="gpt-4.1", stream=True), tools=[WebScraperTool(off_prompt=True), PromptSummaryTool(off_prompt=False)], ) ) pipeline.run()
[02/27/25 20:28:26] INFO PromptTask b15b3f6e25464da0a43e460c82a49861
Input: Based on https://griptape.ai, tell me what
griptape is.
WebScraperTool.get_content (call_pcIDNi1AXKesObKAJs9AK1kc){"values":{"url":"https://griptape.ai"}}[02/27/25 20:28:28] INFO Subtask 4f549b191fe147d2a0d05b626c09c43d
Actions: [
{
"tag": "call_pcIDNi1AXKesObKAJs9AK1kc",
"name": "WebScraperTool",
"path": "get_content",
"input": {
"values": {
"url": "https://griptape.ai"
}
}
}
]
[02/27/25 20:28:32] INFO Subtask 4f549b191fe147d2a0d05b626c09c43d
Response: Output of "WebScraperTool.get_content"
was stored in memory with memory_name "TaskMemory"
and artifact_namespace
"a11e816b9ade4c10b94ed4986ec1982b"
PromptSummaryTool.summarize (call_uLwOt80tnwydh0N433pEKxgk){"values":{"summary":{"memory_name":"TaskMemory","artifact_namespace":"a11e816b9ade4c10b94ed4986ec1982b"}}}[02/27/25 20:28:33] INFO Subtask 57a2be8d678e429d8546f43aff2d2e50
Actions: [
{
"tag": "call_uLwOt80tnwydh0N433pEKxgk",
"name": "PromptSummaryTool",
"path": "summarize",
"input": {
"values": {
"summary": {
"memory_name": "TaskMemory",
"artifact_namespace":
"a11e816b9ade4c10b94ed4986ec1982b"
}
}
}
}
]
[02/27/25 20:28:37] INFO Subtask 57a2be8d678e429d8546f43aff2d2e50
Response: Griptape provides a comprehensive
solution for developers to build, deploy, and scale
AI-powered applications. It offers an open-source
AI framework and a cloud execution runtime,
enabling developers to create business logic using
Python without relying on prompt engineering.
Griptape supports the deployment of ETL, RAG, and
other structures with simple API abstractions,
eliminating the need for infrastructure management.
It also provides tools for monitoring performance
and enforcing policies. The Griptape AI Framework
allows for the creation of Gen AI Agents,
pipelines, and workflows, while the Griptape AI
Cloud handles infrastructure, offering automated
data preparation and retrieval services.
Griptape provides a comprehensive solution for developers to build, deploy, and scale AI-powered applications. It offers an open-source AI framework and a cloud execution runtime, enabling developers to create business logic using Python without relying on prompt engineering. Griptape supports the deployment of ETL, RAG, and other structures with simple API abstractions, eliminating the need for infrastructure management. It also provides tools for monitoring performance and enforcing policies. The Griptape AI Framework allows for the creation of Gen AI Agents, pipelines, and workflows, while the Griptape AI Cloud handles infrastructure, offering automated data preparation and retrieval services.[02/27/25 20:28:39] INFO PromptTask b15b3f6e25464da0a43e460c82a49861
Output: Griptape provides a comprehensive solution
for developers to build, deploy, and scale
AI-powered applications. It offers an open-source
AI framework and a cloud execution runtime,
enabling developers to create business logic using
Python without relying on prompt engineering.
Griptape supports the deployment of ETL, RAG, and
other structures with simple API abstractions,
eliminating the need for infrastructure management.
It also provides tools for monitoring performance
and enforcing policies. The Griptape AI Framework
allows for the creation of Gen AI Agents,
pipelines, and workflows, while the Griptape AI
Cloud handles infrastructure, offering automated
data preparation and retrieval services.
You can also use the TextChunkEvent and ActionChunkEvent to further differentiate the different types of chunks for more customized output.
from griptape.drivers.prompt.openai import OpenAiChatPromptDriver from griptape.events import ActionChunkEvent, EventBus, EventListener, TextChunkEvent from griptape.structures import Pipeline from griptape.tasks import PromptTask from griptape.tools import PromptSummaryTool, WebScraperTool EventBus.add_event_listeners( [ EventListener( lambda e: print(str(e), end="", flush=True), event_types=[TextChunkEvent], ), EventListener( lambda e: print(str(e), end="", flush=True), event_types=[ActionChunkEvent], ), ] ) pipeline = Pipeline() pipeline.add_tasks( PromptTask( "Based on https://griptape.ai, tell me what griptape is.", prompt_driver=OpenAiChatPromptDriver(model="gpt-4.1", stream=True), tools=[WebScraperTool(off_prompt=True), PromptSummaryTool(off_prompt=False)], ) ) pipeline.run()
[02/27/25 20:28:28] INFO PromptTask f28262aaab014f548c8d55f0b0b8fc37
Input: Based on https://griptape.ai, tell me what
griptape is.
WebScraperTool.get_content (call_dpLUPLRdEVuhBbBrTQCtjGDX){"values":{"url":"https://griptape.ai"}}[02/27/25 20:28:30] INFO Subtask fcd088bbcc494d1584a5d5d5004b1e04
Actions: [
{
"tag": "call_dpLUPLRdEVuhBbBrTQCtjGDX",
"name": "WebScraperTool",
"path": "get_content",
"input": {
"values": {
"url": "https://griptape.ai"
}
}
}
]
[02/27/25 20:28:31] INFO Subtask fcd088bbcc494d1584a5d5d5004b1e04
Response: Output of "WebScraperTool.get_content"
was stored in memory with memory_name "TaskMemory"
and artifact_namespace
"bd6dee4ddcb640c1bcb50551d57502a7"
PromptSummaryTool.summarize (call_TGCW6j9ORjJm42ae1mIG4GBb){"values":{"summary":{"memory_name":"TaskMemory","artifact_namespace":"bd6dee4ddcb640c1bcb50551d57502a7"}}}[02/27/25 20:28:32] INFO Subtask fca257f69cbf466e8aa0bcff9e25d79b
Actions: [
{
"tag": "call_TGCW6j9ORjJm42ae1mIG4GBb",
"name": "PromptSummaryTool",
"path": "summarize",
"input": {
"values": {
"summary": {
"memory_name": "TaskMemory",
"artifact_namespace":
"bd6dee4ddcb640c1bcb50551d57502a7"
}
}
}
}
]
[02/27/25 20:28:36] INFO Subtask fca257f69cbf466e8aa0bcff9e25d79b
Response: Griptape offers a comprehensive solution
for developers to build, deploy, and scale
AI-powered applications. It provides an open-source
AI framework and a cloud execution runtime,
enabling developers to create business logic using
Python without relying on prompt engineering.
Griptape's platform supports the development of ETL
pipelines, retrieval patterns, and AI agents, while
offering secure, efficient, and cost-effective
deployment. The Griptape AI Cloud handles
infrastructure management, providing automated data
preparation, retrieval as a service, and runtime
for AI agents and workflows. This allows developers
to focus on building and scaling applications
without the complexities of infrastructure
management.
Griptape is a comprehensive solution designed for developers to build, deploy, and scale AI-powered applications. It offers an open-source AI framework and a cloud execution runtime, allowing developers to create business logic using Python without the need for prompt engineering. The platform supports the development of ETL pipelines, retrieval patterns, and AI agents, providing secure, efficient, and cost-effective deployment options. Griptape AI Cloud manages infrastructure, offering automated data preparation, retrieval as a service, and runtime for AI agents and workflows, enabling developers to focus on application development and scaling without dealing with infrastructure complexities.[02/27/25 20:28:38] INFO PromptTask f28262aaab014f548c8d55f0b0b8fc37
Output: Griptape is a comprehensive solution
designed for developers to build, deploy, and scale
AI-powered applications. It offers an open-source
AI framework and a cloud execution runtime,
allowing developers to create business logic using
Python without the need for prompt engineering. The
platform supports the development of ETL pipelines,
retrieval patterns, and AI agents, providing
secure, efficient, and cost-effective deployment
options. Griptape AI Cloud manages infrastructure,
offering automated data preparation, retrieval as a
service, and runtime for AI agents and workflows,
enabling developers to focus on application
development and scaling without dealing with
infrastructure complexities.
If you want Gen AI Builder to handle the chunk events for you, use the Stream utility to automatically wrap BaseChunkEvents in a Python iterator.
The Stream utility does not automatically enable streaming on the Drivers that produce BaseChunkEvents.
Make sure to enable streaming on the Drivers or else Stream will yield no iterations.
import logging from griptape.configs import Defaults from griptape.structures import Agent from griptape.tools import PromptSummaryTool, WebScraperTool from griptape.utils import Stream # Hide Griptape's usual output logging.getLogger(Defaults.logging_config.logger_name).setLevel(logging.ERROR) agent = Agent( input="Based on https://griptape.ai, tell me what griptape is.", tools=[ PromptSummaryTool(off_prompt=True), WebScraperTool(off_prompt=False), ], stream=True, ) for artifact in Stream(agent).run(): print(artifact.value, end="", flush=True)
WebScraperTool.call_XNlfty9CYNnlkIcKMFo2USPw (get_content){
"values": {
"url": "https://griptape.ai"
}
}
Griptape is a platform designed to help developers build, deploy, and scale end-to-end solutions, particularly those powered by large language models (LLMs). It offers tools for data preparation, retrieval, AI agents, pipelines, and workflows. Griptape provides an open-source AI framework and an execution runtime called Griptape AI Cloud.
Key features of Griptape include:
- **Build & Secure**: Developers can build business logic using Python, enhancing security, performance, and reducing costs with Off-Prompt™ technology.
- **Deploy & Scale**: It allows for the deployment and running of ETL, RAG, and other structures without managing infrastructure, enabling seamless scaling.
- **Manage & Monitor**: Users can monitor performance, reliability, and spending, and enforce policies across the organization.
Griptape AI Framework offers abstractions for building AI agents, systems of agents, pipelines, workflows, and retrieval-augmented generation (RAG) implementations without needing extensive knowledge of Gen AI or prompt engineering.
Griptape AI Cloud handles infrastructure management, hosting everything from data processing pipelines to serverless application runtimes. It includes features like automated data preparation (ETL), retrieval as a service (RAG), and structure runtime (RUN) for building AI agents and workflows.
Sometimes, streaming can be too verbose. You can use Stream.event_types to only listen to specific event types. A good example is to remove the ActionChunkEvent from the stream if you don't need to see events related to Tool usage.
import logging from griptape.configs import Defaults from griptape.events import FinishPromptEvent, FinishStructureRunEvent, TextChunkEvent from griptape.structures import Agent from griptape.tools import PromptSummaryTool, WebScraperTool from griptape.utils import Stream # Hide Griptape's usual output logging.getLogger(Defaults.logging_config.logger_name).setLevel(logging.ERROR) agent = Agent( input="Based on https://griptape.ai, tell me what griptape is.", tools=[ PromptSummaryTool(off_prompt=True), WebScraperTool(off_prompt=False), ], stream=True, ) # Listen for the following event types event_types = [TextChunkEvent, FinishPromptEvent, FinishStructureRunEvent] for artifact in Stream(agent, event_types=event_types).run(): print(artifact.value, end="", flush=True)
--8<-- "docs/griptape-framework/misc/logs/events_4_filtered.txt"
Counting Tokens
To count tokens, you can use Event Listeners and the TokenCounter util:
--8<-- "docs/griptape-framework/misc/src/events_5.py"
--8<-- "docs/griptape-framework/misc/logs/events_5.txt"
Inspecting Payloads
You can use the StartPromptEvent to inspect the Prompt Stack and final prompt string before it is sent to the LLM.
--8<-- "docs/griptape-framework/misc/src/events_6.py"
--8<-- "docs/griptape-framework/misc/logs/events_6.txt"
... Prompt Stack Messages: system: user: Write me a poem. Final Prompt String: User: Write me a poem. Assistant: ...
EventListenerDriver.on_event Return Value Behavior
The value that gets returned from the EventListener.on_event will determine what gets sent to the event_listener_driver.
EventListener.on_event is None
By default, the EventListener.on_event function is None. Any events that the EventListener is listening for will get sent to the event_listener_driver as-is.
Return BaseEvent or dict
You can return a BaseEvent or dict object from EventListener.on_event, and it will get sent to the event_listener_driver.
Return None
You can return None in the on_event function to prevent the event from getting sent to the event_listener_driver.
--8<-- "docs/griptape-framework/misc/src/events_no_publish.py"
--8<-- "docs/griptape-framework/misc/logs/events_no_publish.txt"
← Prev
Eval Engine
↑ Up
Gen AI Builder: A Python Framework for Building Generative AI Applications
Next →
Tokenizers
Could this page be better? Report a problem or suggest an addition!