@aws/agentcore 0.3.0-preview.1.0

package/dist/assets/python/crewai/base/gitignore.template

@@ -0,0 +1,40 @@
+# Environment variables
+.env
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments
+venv/
+ENV/
+env/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db

package/dist/assets/python/crewai/base/main.py

@@ -0,0 +1,55 @@
+from crewai import Agent, Crew, Task, Process
+from crewai.tools import tool
+from bedrock_agentcore.runtime import BedrockAgentCoreApp
+from model.load import load_model
+
+app = BedrockAgentCoreApp()
+log = app.logger
+
+
+# Define a simple function tool
+@tool
+def add_numbers(a: int, b: int) -> int:
+    """Return the sum of two numbers"""
+    return a + b
+
+
+# Define a collection of tools used by the model
+tools = [add_numbers]
+
+
+@app.entrypoint
+def invoke(payload, context):
+    log.info("Invoking Agent.....")
+
+    # Define the Agent with Tools
+    agent = Agent(
+        role="Question Answering Assistant",
+        goal="Answer the users questions",
+        backstory="Always eager to answer any questions",
+        llm=load_model(),
+        tools=tools,
+    )
+
+    # Define the Task
+    task = Task(
+        agent=agent,
+        description="Answer the users question: {prompt}",
+        expected_output="An answer to the users question",
+    )
+
+    # Create the Crew
+    crew = Crew(agents=[agent], tasks=[task], process=Process.sequential)
+
+    # Process the user prompt
+    prompt = payload.get("prompt", "What can you help me with?")
+
+    # Run the crew
+    result = crew.kickoff(inputs={"prompt": prompt})
+
+    # Return result
+    return {"result": result.raw}
+
+
+if __name__ == "__main__":
+    app.run()

package/dist/assets/python/crewai/base/model/load.py

@@ -0,0 +1,133 @@
+{{#if (eq modelProvider "Bedrock")}}
+from crewai import LLM
+
+# Uses global inference profile for Claude Sonnet 4.5
+# https://docs.aws.amazon.com/bedrock/latest/userguide/inference-profiles-support.html
+MODEL_ID = "bedrock/global.anthropic.claude-sonnet-4-5-20250929-v1:0"
+
+
+def load_model() -> LLM:
+    """Get Bedrock model client using IAM credentials."""
+    return LLM(model=MODEL_ID)
+{{/if}}
+{{#if (eq modelProvider "Anthropic")}}
+import os
+from crewai import LLM
+from bedrock_agentcore.identity.auth import requires_api_key
+
+IDENTITY_PROVIDER_NAME = "{{identityProviders.[0].name}}"
+IDENTITY_ENV_VAR = "{{identityProviders.[0].envVarName}}"
+
+
+@requires_api_key(provider_name=IDENTITY_PROVIDER_NAME)
+def _agentcore_identity_api_key_provider(api_key: str) -> str:
+    """Fetch API key from AgentCore Identity."""
+    return api_key
+
+
+def _get_api_key() -> str:
+    """
+    Uses AgentCore Identity for API key management in deployed environments.
+    For local development, run via 'agentcore dev' which loads agentcore/.env.
+    """
+    if os.getenv("LOCAL_DEV") == "1":
+        api_key = os.getenv(IDENTITY_ENV_VAR)
+        if not api_key:
+            raise RuntimeError(
+                f"{IDENTITY_ENV_VAR} not found. Add {IDENTITY_ENV_VAR}=your-key to .env.local"
+            )
+        return api_key
+    return _agentcore_identity_api_key_provider()
+
+
+def load_model() -> LLM:
+    """Get authenticated Anthropic model client."""
+    api_key = _get_api_key()
+    # CrewAI requires ANTHROPIC_API_KEY env var (ignores api_key parameter)
+    os.environ["ANTHROPIC_API_KEY"] = api_key
+    return LLM(
+        model="anthropic/claude-sonnet-4-5-20250929",
+        api_key=api_key,
+        max_tokens=4096
+    )
+{{/if}}
+{{#if (eq modelProvider "OpenAI")}}
+import os
+from crewai import LLM
+from bedrock_agentcore.identity.auth import requires_api_key
+
+IDENTITY_PROVIDER_NAME = "{{identityProviders.[0].name}}"
+IDENTITY_ENV_VAR = "{{identityProviders.[0].envVarName}}"
+
+
+@requires_api_key(provider_name=IDENTITY_PROVIDER_NAME)
+def _agentcore_identity_api_key_provider(api_key: str) -> str:
+    """Fetch API key from AgentCore Identity."""
+    return api_key
+
+
+def _get_api_key() -> str:
+    """
+    Uses AgentCore Identity for API key management in deployed environments.
+    For local development, run via 'agentcore dev' which loads agentcore/.env.
+    """
+    if os.getenv("LOCAL_DEV") == "1":
+        api_key = os.getenv(IDENTITY_ENV_VAR)
+        if not api_key:
+            raise RuntimeError(
+                f"{IDENTITY_ENV_VAR} not found. Add {IDENTITY_ENV_VAR}=your-key to .env.local"
+            )
+        return api_key
+    return _agentcore_identity_api_key_provider()
+
+
+def load_model() -> LLM:
+    """Get authenticated OpenAI model client."""
+    api_key = _get_api_key()
+    # CrewAI requires OPENAI_API_KEY env var (ignores api_key parameter)
+    os.environ["OPENAI_API_KEY"] = api_key
+    return LLM(
+        model="openai/gpt-4o",
+        api_key=api_key
+    )
+{{/if}}
+{{#if (eq modelProvider "Gemini")}}
+import os
+from crewai import LLM
+from bedrock_agentcore.identity.auth import requires_api_key
+
+IDENTITY_PROVIDER_NAME = "{{identityProviders.[0].name}}"
+IDENTITY_ENV_VAR = "{{identityProviders.[0].envVarName}}"
+
+
+@requires_api_key(provider_name=IDENTITY_PROVIDER_NAME)
+def _agentcore_identity_api_key_provider(api_key: str) -> str:
+    """Fetch API key from AgentCore Identity."""
+    return api_key
+
+
+def _get_api_key() -> str:
+    """
+    Uses AgentCore Identity for API key management in deployed environments.
+    For local development, run via 'agentcore dev' which loads agentcore/.env.
+    """
+    if os.getenv("LOCAL_DEV") == "1":
+        api_key = os.getenv(IDENTITY_ENV_VAR)
+        if not api_key:
+            raise RuntimeError(
+                f"{IDENTITY_ENV_VAR} not found. Add {IDENTITY_ENV_VAR}=your-key to .env.local"
+            )
+        return api_key
+    return _agentcore_identity_api_key_provider()
+
+
+def load_model() -> LLM:
+    """Get authenticated Gemini model client."""
+    api_key = _get_api_key()
+    # CrewAI requires GEMINI_API_KEY env var (ignores api_key parameter)
+    os.environ["GEMINI_API_KEY"] = api_key
+    return LLM(
+        model="gemini/gemini-2.0-flash",
+        api_key=api_key
+    )
+{{/if}}

package/dist/assets/python/crewai/base/pyproject.toml

@@ -0,0 +1,32 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "{{ name }}"
+version = "0.1.0"
+description = "AgentCore Runtime Application using CrewAI SDK"
+readme = "README.md"
+requires-python = ">=3.11"
+dependencies = [
+    "opentelemetry-distro",
+    "opentelemetry-exporter-otlp",
+    "bedrock-agentcore >= 1.0.3",
+    "botocore[crt] >= 1.35.0",
+    "python-dotenv >= 1.0.1",
+{{#if (eq modelProvider "Bedrock")}}
+    "crewai[tools,bedrock] >= 1.3.0",
+{{/if}}
+{{#if (eq modelProvider "Anthropic")}}
+    "crewai[tools,anthropic] >= 1.3.0",
+{{/if}}
+{{#if (eq modelProvider "OpenAI")}}
+    "crewai[tools,openai] >= 1.3.0",
+{{/if}}
+{{#if (eq modelProvider "Gemini")}}
+    "crewai[tools,google-genai] >= 1.3.0",
+{{/if}}
+]
+
+[tool.hatch.build.targets.wheel]
+packages = ["."]

package/dist/assets/python/googleadk/base/README.md

@@ -0,0 +1,39 @@
+This is a project generated by the agentcore create CLI tool!
+
+# Layout
+
+There is one directory with generated application code, `src/` . At the root, there is a `.gitignore` file, a
+`.agentcore` folder which represents the configurations and state associated with this project. Other `agentcore`
+commands like `deploy`, `dev`, and `invoke` rely on the configuration stored here.
+
+## src/
+
+The main entrypoint to your app is defined in `src/main.py`. Using the AgentCore SDK `@app.entrypoint` decorator, this
+file defines a Starlette ASGI app with the Google ADK framework running within.
+
+`src/model/load.py` instantiates your chosen model provider (Gemini).
+
+## Environment Variables
+
+| Variable                    | Required | Description                                                      |
+| --------------------------- | -------- | ---------------------------------------------------------------- |
+| `AGENTCORE_IDENTITY_GEMINI` | Yes      | Gemini API key (local) or Identity provider name (deployed)      |
+| `LOCAL_DEV`                 | No       | Set to `1` to use `agentcore/.env` instead of AgentCore Identity |
+
+# Developing locally
+
+If installation was successful, a virtual environment is already created with dependencies installed.
+
+Run `source .venv/bin/activate` before developing.
+
+`agentcore dev` will start a local server on 0.0.0.0:8080.
+
+In a new terminal, you can invoke that server with:
+
+`agentcore invoke --dev "What can you do"`
+
+# Deployment
+
+After providing credentials, `agentcore deploy` will deploy your project into Amazon Bedrock AgentCore.
+
+Use `agentcore invoke` to invoke your deployed agent.

package/dist/assets/python/googleadk/base/gitignore.template

@@ -0,0 +1,40 @@
+# Environment variables
+.env
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments
+venv/
+ENV/
+env/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db

package/dist/assets/python/googleadk/base/main.py

@@ -0,0 +1,84 @@
+import os
+from google.adk.agents import Agent
+from google.adk.runners import Runner
+from google.adk.sessions import InMemorySessionService
+from google.genai import types
+from bedrock_agentcore.runtime import BedrockAgentCoreApp
+from model.load import load_model
+from mcp_client.client import get_streamable_http_mcp_client
+
+app = BedrockAgentCoreApp()
+log = app.logger
+
+APP_NAME = "{{ name }}"
+
+# https://google.github.io/adk-docs/agents/models/
+MODEL_ID = "gemini-2.5-flash"
+
+
+# Define a simple function tool
+def add_numbers(a: int, b: int) -> int:
+    """Return the sum of two numbers"""
+    return a + b
+
+
+# Set environment variables for model authentication
+load_model()
+
+# Get MCP Toolset
+mcp_toolset = [get_streamable_http_mcp_client()]
+
+# Agent Definition
+agent = Agent(
+    model=MODEL_ID,
+    name="{{ name }}",
+    description="Agent to answer questions",
+    instruction="I can answer your questions using the knowledge I have!",
+    tools=mcp_toolset + [add_numbers],
+)
+
+
+# Session and Runner
+async def setup_session_and_runner(user_id, session_id):
+    session_service = InMemorySessionService()
+    session = await session_service.create_session(
+        app_name=APP_NAME, user_id=user_id, session_id=session_id
+    )
+    runner = Runner(agent=agent, app_name=APP_NAME, session_service=session_service)
+    return session, runner
+
+
+# Agent Interaction
+async def call_agent_async(query, user_id, session_id):
+    content = types.Content(role="user", parts=[types.Part(text=query)])
+    session, runner = await setup_session_and_runner(user_id, session_id)
+    events = runner.run_async(
+        user_id=user_id, session_id=session.id, new_message=content
+    )
+
+    final_response = None
+    async for event in events:
+        if event.is_final_response():
+            final_response = event.content.parts[0].text
+
+    return final_response
+
+
+@app.entrypoint
+async def invoke(payload, context):
+    log.info("Invoking Agent.....")
+
+    # Process the user prompt
+    prompt = payload.get("prompt", "What can you help me with?")
+    session_id = getattr(context, "session_id", "default_session")
+    user_id = payload.get("user_id", "default_user")
+
+    # Run the agent
+    result = await call_agent_async(prompt, user_id, session_id)
+
+    # Return result
+    return {"result": result}
+
+
+if __name__ == "__main__":
+    app.run()

package/dist/assets/python/googleadk/base/mcp_client/client.py

@@ -0,0 +1,15 @@
+from google.adk.tools.mcp_tool.mcp_toolset import MCPToolset
+from google.adk.tools.mcp_tool.mcp_session_manager import StreamableHTTPConnectionParams
+
+# ExaAI provides information about code through web searches, crawling and code context searches through their platform. Requires no authentication
+EXAMPLE_MCP_ENDPOINT = "https://mcp.exa.ai/mcp"
+
+
+def get_streamable_http_mcp_client() -> MCPToolset:
+    """
+    Returns an MCP Toolset compatible with Google ADK.
+    """
+    # to use an MCP server that supports bearer authentication, add headers={"Authorization": f"Bearer {access_token}"}
+    return MCPToolset(
+        connection_params=StreamableHTTPConnectionParams(url=EXAMPLE_MCP_ENDPOINT)
+    )

package/dist/assets/python/googleadk/base/model/load.py

@@ -0,0 +1,41 @@
+import os
+from bedrock_agentcore.identity.auth import requires_api_key
+
+IDENTITY_PROVIDER_NAME = "{{identityProviders.[0].name}}"
+IDENTITY_ENV_VAR = "{{identityProviders.[0].envVarName}}"
+
+
+@requires_api_key(provider_name=IDENTITY_PROVIDER_NAME)
+def _agentcore_identity_api_key_provider(api_key: str) -> str:
+    """Fetch API key from AgentCore Identity."""
+    return api_key
+
+
+def _get_api_key() -> str:
+    """
+    Uses AgentCore Identity for API key management in deployed environments.
+    For local development, run via 'agentcore dev' which loads agentcore/.env.
+    """
+    if os.getenv("LOCAL_DEV") == "1":
+        api_key = os.getenv(IDENTITY_ENV_VAR)
+        if not api_key:
+            raise RuntimeError(
+                f"{IDENTITY_ENV_VAR} not found. Add {IDENTITY_ENV_VAR}=your-key to .env.local"
+            )
+        return api_key
+    return _agentcore_identity_api_key_provider()
+
+
+def load_model() -> None:
+    """
+    Set up Gemini API key authentication.
+    Uses AgentCore Identity for API key management in deployed environments,
+    and falls back to .env file for local development.
+    Sets the GOOGLE_API_KEY environment variable for the Google ADK.
+    """
+    api_key = _get_api_key()
+    # Use Google AI Studios API Key Authentication.
+    # https://google.github.io/adk-docs/agents/models/#google-ai-studio
+    os.environ["GOOGLE_API_KEY"] = api_key
+    # Set to TRUE is using Google Vertex AI, Set to FALSE for Google AI Studio
+    os.environ["GOOGLE_GENAI_USE_VERTEXAI"] = "FALSE"

package/dist/assets/python/googleadk/base/pyproject.toml

@@ -0,0 +1,21 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "{{ name }}"
+version = "0.1.0"
+description = "AgentCore Runtime Application using Google ADK"
+readme = "README.md"
+requires-python = ">=3.10"
+dependencies = [
+    "opentelemetry-distro",
+    "opentelemetry-exporter-otlp",
+    "google-adk >= 1.17.0",
+    "bedrock-agentcore >= 1.0.3",
+    "botocore[crt] >= 1.35.0",
+    "python-dotenv >= 1.0.1",
+]
+
+[tool.hatch.build.targets.wheel]
+packages = ["."]

package/dist/assets/python/langchain_langgraph/base/README.md

@@ -0,0 +1,41 @@
+This is a project generated by the agentcore create CLI tool!
+
+# Layout
+
+There is one directory with generated application code, `src/` . At the root, there is a `.gitignore` file, a
+`.agentcore` folder which represents the configurations and state associated with this project. Other `agentcore`
+commands like `deploy`, `dev`, and `invoke` rely on the configuration stored here.
+
+## src/
+
+The main entrypoint to your app is defined in `src/main.py`. Using the AgentCore SDK `@app.entrypoint` decorator, this
+file defines a Starlette ASGI app with the LangChain/LangGraph framework running within.
+
+`src/model/load.py` instantiates your chosen model provider.
+
+## Environment Variables
+
+| Variable                       | Required        | Description                                                      |
+| ------------------------------ | --------------- | ---------------------------------------------------------------- |
+| `AGENTCORE_IDENTITY_OPENAI`    | Yes (OpenAI)    | OpenAI API key (local) or Identity provider name (deployed)      |
+| `AGENTCORE_IDENTITY_ANTHROPIC` | Yes (Anthropic) | Anthropic API key (local) or Identity provider name (deployed)   |
+| `AGENTCORE_IDENTITY_GEMINI`    | Yes (Gemini)    | Gemini API key (local) or Identity provider name (deployed)      |
+| `LOCAL_DEV`                    | No              | Set to `1` to use `agentcore/.env` instead of AgentCore Identity |
+
+# Developing locally
+
+If installation was successful, a virtual environment is already created with dependencies installed.
+
+Run `source .venv/bin/activate` before developing.
+
+`agentcore dev` will start a local server on 0.0.0.0:8080.
+
+In a new terminal, you can invoke that server with:
+
+`agentcore invoke --dev "What can you do"`
+
+# Deployment
+
+After providing credentials, `agentcore deploy` will deploy your project into Amazon Bedrock AgentCore.
+
+Use `agentcore invoke` to invoke your deployed agent.

package/dist/assets/python/langchain_langgraph/base/gitignore.template

@@ -0,0 +1,40 @@
+# Environment variables
+.env
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments
+venv/
+ENV/
+env/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db

package/dist/assets/python/langchain_langgraph/base/main.py

@@ -0,0 +1,51 @@
+import os
+from langchain_core.messages import HumanMessage
+from langgraph.prebuilt import create_react_agent
+from langchain.tools import tool
+from bedrock_agentcore.runtime import BedrockAgentCoreApp
+from model.load import load_model
+from mcp_client.client import get_streamable_http_mcp_client
+
+app = BedrockAgentCoreApp()
+log = app.logger
+
+# Instantiate model
+llm = load_model()
+
+
+# Define a simple function tool
+@tool
+def add_numbers(a: int, b: int) -> int:
+    """Return the sum of two numbers"""
+    return a + b
+
+
+# Define a collection of tools used by the model
+tools = [add_numbers]
+
+
+@app.entrypoint
+async def invoke(payload, context):
+    log.info("Invoking Agent.....")
+
+    # Get MCP Client
+    mcp_client = get_streamable_http_mcp_client()
+
+    # Load MCP Tools
+    mcp_tools = await mcp_client.get_tools()
+
+    # Define the agent using create_react_agent
+    graph = create_react_agent(llm, tools=mcp_tools + tools)
+
+    # Process the user prompt
+    prompt = payload.get("prompt", "What can you help me with?")
+
+    # Run the agent
+    result = await graph.ainvoke({"messages": [HumanMessage(content=prompt)]})
+
+    # Return result
+    return {"result": result["messages"][-1].content}
+
+
+if __name__ == "__main__":
+    app.run()

package/dist/assets/python/langchain_langgraph/base/mcp_client/client.py

@@ -0,0 +1,19 @@
+from langchain_mcp_adapters.client import MultiServerMCPClient
+
+# ExaAI provides information about code through web searches, crawling and code context searches through their platform. Requires no authentication
+EXAMPLE_MCP_ENDPOINT = "https://mcp.exa.ai/mcp"
+
+
+def get_streamable_http_mcp_client() -> MultiServerMCPClient:
+    """
+    Returns an MCP Client compatible with LangChain/LangGraph.
+    """
+    # to use an MCP server that supports bearer authentication, add headers={"Authorization": f"Bearer {access_token}"}
+    return MultiServerMCPClient(
+        {
+            "agentcore_gateway": {
+                "transport": "streamable_http",
+                "url": EXAMPLE_MCP_ENDPOINT,
+            }
+        }
+    )