feat: Autogen AI framework extension

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())

python/README.md

@@ -34,6 +34,12 @@ PydanticAI extension of AgentKit. Enables agentic workflows to interact with onc
 
 See [AgentKit PydanticAI](./framework-extensions/pydantic-ai/README.md) to get started!
 
+#### `coinbase-agentkit-autogen`
+
+Microsoft Autogen extension of AgentKit. Enables agentic workflows to interact with onchain actions.
+
+See [AgentKit Autogen](./framework-extensions/autogen/README.md)
+
 ### `create-onchain-agent`
 
 A quickstart CLI tool to scaffold out a chatbot using Coinbase AgentKit. Runnable via `pipx run create-onchain-agent`.

python/examples/autogen-cdp-chatbot/.env.local

@@ -0,0 +1,8 @@
+# Required
+CDP_API_KEY_ID=
+CDP_API_KEY_SECRET=
+CDP_WALLET_SECRET=
+OPENAI_API_KEY=
+
+# Optional with defaults
+NETWORK_ID=base-sepolia

python/examples/autogen-cdp-chatbot/Makefile

@@ -0,0 +1,29 @@
+ifneq (,$(wildcard ./.env))
+	include .env
+endif
+
+export
+
+.PHONY: install
+install:
+	uv sync
+
+.PHONY: run
+run:
+	uv run chatbot.py
+
+.PHONY: format
+format:
+	uv run ruff format .
+
+.PHONY: format-check
+format-check:
+	uv run ruff format . --check
+
+.PHONY: lint
+lint:
+	uv run ruff check .
+
+.PHONY: lint-fix
+lint-fix:
+	uv run ruff check . --fix

python/examples/autogen-cdp-chatbot/README.md

@@ -0,0 +1,50 @@
+# CDP Agentkit Autogen Extension Examples - Chatbot Python
+
+This example demonstrates an agent setup as a terminal style chatbot using Microsoft Autogen integration with CDP Agentkit actions.
+
+## Ask the chatbot to engage in the Web3 ecosystem!
+
+- "What is the price of BTC?"
+- "Request some test tokens from the faucet"
+- "What are my CDP account details?"
+- "Transfer a portion of your ETH to a random address"
+- "Deploy an NFT that will go super viral!"
+- "Deploy an ERC-20 token with total supply 1 billion"
+
+## Requirements
+
+- Python 3.10+
+- uv for package management and tooling
+ - [uv Installation Instructions](https://github.com/astral-sh/uv?tab=readme-ov-file#installation)
+- [CDP API Key](https://portal.cdp.coinbase.com/access/api)
+- [OpenAI API Key](https://platform.openai.com/docs/quickstart#create-and-export-an-api-key)
+
+### Checking Python Version
+
+Before using the example, ensure that you have the correct version of Python installed. The example requires Python 3.10 or higher. You can check your Python version by running:
+
+```bash
+python --version
+uv --version
+```
+
+## Installation
+
+```bash
+make install
+```
+
+## Run the Chatbot
+
+### Env
+
+Rename .env.local to .env and ensure the following var are set:
+
+- "CDP_API_KEY_ID"
+- "CDP_API_KEY_SECRET"
+- "CDP_WALLET_SECRET" # This is the wallet secret for the wallet you want to use for the chatbot
+- "OPENAI_API_KEY"
+
+```bash
+make run
+```

python/examples/autogen-cdp-chatbot/chatbot.py

@@ -0,0 +1,270 @@
+import asyncio
+import json
+import os
+import time
+
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.messages import BaseChatMessage, TextMessage
+from autogen_core.memory import ListMemory
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from coinbase_agentkit import (
+ AgentKit,
+ AgentKitConfig,
+ CdpEvmWalletProvider,
+ CdpEvmWalletProviderConfig,
+ cdp_api_action_provider,
+ erc20_action_provider,
+ pyth_action_provider,
+ wallet_action_provider,
+ weth_action_provider,
+ x402_action_provider,
+)
+from coinbase_agentkit_autogen.autogen_tools import get_autogen_tools
+from dotenv import load_dotenv
+
+
+def initialize_agent(
+ config: CdpEvmWalletProviderConfig,
+) -> tuple[AssistantAgent, CdpEvmWalletProvider]:
+ """Initialize the agent with CDP Agentkit.
+
+ Args:
+ config: Configuration for the CDP EVM Server Wallet Provider
+
+ Returns:
+ tuple[Agent, CdpEvmServerWalletProvider]: The initialized agent and wallet provider
+
+ """
+ # Initialize the language model
+ model_client = OpenAIChatCompletionClient(
+ model="gpt-4o-mini",
+ )
+
+ # Initialize the wallet provider with the config
+ wallet_provider = CdpEvmWalletProvider(
+ CdpEvmWalletProviderConfig(
+ api_key_id=config.api_key_id, # CDP API Key ID
+ api_key_secret=config.api_key_secret, # CDP API Key Secret
+ wallet_secret=config.wallet_secret, # CDP Wallet Secret
+ network_id=config.network_id, # Network ID - Optional, will default to 'base-sepolia'
+ address=config.address, # Wallet Address - Optional, will trigger idempotency flow if not provided
+ idempotency_key=config.idempotency_key, # Idempotency Key - Optional, seeds generation of a new wallet
+ rpc_url=config.rpc_url, # Optional RPC URL override
+ )
+ )
+
+ # Create AgentKit instance with wallet and action providers
+ agentkit = AgentKit(
+ AgentKitConfig(
+ wallet_provider=wallet_provider,
+ action_providers=[
+ cdp_api_action_provider(),
+ erc20_action_provider(),
+ pyth_action_provider(),
+ wallet_action_provider(),
+ weth_action_provider(),
+ x402_action_provider(),
+ ],
+ )
+ )
+
+ # Get tools for the agent
+ tools = get_autogen_tools(agentkit)
+
+ # Set up conversation memory
+ memory = ListMemory()
+
+ # Set up Autogen Assistant Agent
+ agent = AssistantAgent(
+ name="assistant",
+ model_client=model_client,
+ reflect_on_tool_use=True,
+ tools=tools,
+ memory=[memory],
+ max_tool_iterations=3,
+ system_message=(
+ "You are a helpful agent that can interact onchain using the Coinbase Developer Platform AgentKit. "
+ "You are empowered to interact onchain using your tools. If you ever need funds, you can request "
+ "them from the faucet if you are on network ID 'base-sepolia'. If not, you can provide your wallet "
+ "details and request funds from the user. Before executing your first action, get the wallet details "
+ "to see what network you're on. If there is a 5XX (internal) HTTP error code, ask the user to try "
+ "again later. If someone asks you to do something you can't do with your currently available tools, "
+ "you must say so, and encourage them to implement it themselves using the CDP SDK + Agentkit, "
+ "recommend they go to docs.cdp.coinbase.com for more information. Be concise and helpful with your "
+ "responses. Refrain from restating your tools' descriptions unless it is explicitly requested."
+ ),
+ )
+
+ # Create and return the agent and wallet provider
+ return agent, wallet_provider
+
+
+def setup() -> AssistantAgent:
+ """Set up the agent with persistent wallet storage.
+
+ Returns:
+ AssistantAgent: The initialized agent used for handling wallet interactions and
+ executing associated operations.
+
+ """
+ # Configure network and file path
+ network_id = os.getenv("NETWORK_ID", "base-sepolia")
+ wallet_file = f"wallet_data_{network_id.replace('-', '_')}.txt"
+
+ # Load existing wallet data if available
+ wallet_data = {}
+ if os.path.exists(wallet_file):
+ try:
+ with open(wallet_file) as f:
+ wallet_data = json.load(f)
+ print(f"Loading existing wallet from {wallet_file}")
+ except json.JSONDecodeError:
+ print(f"Warning: Invalid wallet data for {network_id}")
+ wallet_data = {}
+
+ # Determine wallet address using priority order
+ wallet_address = (
+ wallet_data.get("address") # First priority: Saved wallet file
+ or os.getenv("ADDRESS") # Second priority: ADDRESS env var
+ or None # Will trigger idempotency flow if needed
+ )
+
+ # Create the wallet provider config
+ config = CdpEvmWalletProviderConfig(
+ api_key_id=os.getenv("CDP_API_KEY_ID"),
+ api_key_secret=os.getenv("CDP_API_KEY_SECRET"),
+ wallet_secret=os.getenv("CDP_WALLET_SECRET"),
+ network_id=network_id,
+ address=wallet_address,
+ # Only include idempotency_key if we need to create a new wallet
+ idempotency_key=(os.getenv("IDEMPOTENCY_KEY") if not wallet_address else None),
+ )
+
+ # Initialize the agent and get the wallet provider
+ agent_executor, wallet_provider = initialize_agent(config)
+
+ # Save the wallet data after successful initialization
+ new_wallet_data = {
+ "address": wallet_provider.get_address(),
+ "network_id": network_id,
+ "created_at": time.strftime("%Y-%m-%d %H:%M:%S")
+ if not wallet_data
+ else wallet_data.get("created_at"),
+ }
+
+ with open(wallet_file, "w") as f:
+ json.dump(new_wallet_data, f, indent=2)
+ print(f"Wallet data saved to {wallet_file}")
+
+ return agent_executor
+
+
+# Autonomous Mode
+async def run_autonomous_mode(agent_executor: AssistantAgent, interval=10) -> None:
+ """Run the autonomous mode of the AssistantAgent with a specified interval between actions.
+
+ Args:
+ agent_executor (AssistantAgent): The agent responsible for executing tasks autonomously.
+ interval (int, optional): Time in seconds to wait between consecutive autonomous actions. Defaults to 10.
+
+ """
+ print("Starting autonomous mode...")
+ while True:
+ try:
+ # Provide instructions autonomously
+ thought = (
+ "Be creative and do something interesting on the blockchain. "
+ "Choose an action or set of actions and execute it that highlights your abilities."
+ )
+
+ # Run agent in autonomous mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=thought, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ # Wait before the next action
+ await asyncio.sleep(interval)
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Chat Mode
+async def run_chat_mode(agent_executor: AssistantAgent) -> None:
+ """Run the assistant agent in chat mode.
+
+ Args:
+ agent_executor (AssistantAgent): The agent executor responsible for handling the
+ chatbot functionality and running tasks based on user input.
+
+ """
+ print("Starting chat mode... Type 'exit' to end.")
+ while True:
+ try:
+ user_input = input("\nPrompt: ")
+ if user_input.lower() == "exit":
+ break
+
+ # Run agent with the user's input in chat mode
+ async for chunk in agent_executor.run_stream(
+ task=TextMessage(content=user_input, source="User"), output_task_messages=False
+ ):
+ if isinstance(chunk, BaseChatMessage):
+ print("\n-------------------")
+ print(chunk.to_text())
+ print("-------------------")
+
+ except KeyboardInterrupt:
+ print("Goodbye Agent!")
+ await agent_executor._model_client.close()
+ return
+
+
+# Mode Selection
+def choose_mode():
+ """Prompt the user to select a mode of operation from the available options.
+
+ Returns
+ -------
+ str
+ The selected mode, either "chat" or "auto".
+
+ """
+ while True:
+ print("\nAvailable modes:")
+ print("1. chat - Interactive chat mode")
+ print("2. auto - Autonomous action mode")
+
+ choice = input("\nChoose a mode (enter number or name): ").lower().strip()
+ if choice in ["1", "chat"]:
+ return "chat"
+ elif choice in ["2", "auto"]:
+ return "auto"
+ print("Invalid choice. Please try again.")
+
+
+async def main():
+ """Start the chatbot agent."""
+ # Load environment variables
+ load_dotenv()
+
+ # Set up the agent
+ agent_executor = setup()
+
+ # Run the agent in the selected mode
+ mode = choose_mode()
+ if mode == "chat":
+ await run_chat_mode(agent_executor=agent_executor)
+ elif mode == "auto":
+ await run_autonomous_mode(agent_executor=agent_executor)
+
+
+if __name__ == "__main__":
+ print("Starting Agent...")
+ asyncio.run(main())