chore: import upstream snapshot with attribution

acp-code/README.md

@@ -0,0 +1,85 @@
+# Summary Generator multi-agent workflow with ACP
+
+A simple demonstration of the Agent Communication Protocol (ACP), showcasing how two agents built using different frameworks (CrewAI and Smolagents) can collaborate seamlessly to generate and verify a research summary.
+
+---
+
+## Setup and Installation
+
+1. **Install Ollama:**
+   ```bash
+   # Setting up Ollama on linux
+   curl -fsSL https://ollama.com/install.sh | sh
+
+   # Pull the Qwen2.5 model
+   ollama pull qwen2.5:14b
+   ```
+
+2. **Install project dependencies:**
+
+    Ensure you have Python 3.10 or later installed on your system.
+
+    First, install `uv` and set up the environment:
+    ```bash
+    # MacOS/Linux
+    curl -LsSf https://astral.sh/uv/install.sh | sh
+
+    # Windows
+    powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
+    ```
+
+    Install dependencies:
+    ```bash
+    # Create a new directory for our project
+    uv init acp-project
+    cd acp-project
+
+    # Create virtual environment and activate it
+    uv venv
+    source .venv/bin/activate  # MacOS/Linux
+
+    .venv\Scripts\activate     # Windows
+
+    # Install dependencies
+    uv add acp-sdk crewai smolagents duckduckgo-search ollama
+    ```
+
+You can also use any other LLM providers such as OpenAI or Anthropic. Create a `.env` file and add your API keys
+```
+OPENAI_API_KEY=your_openai_key
+ANTHROPIC_API_KEY=your_anthropic_key
+```
+
+## Usage
+Start the two ACP servers in separate terminals:
+
+```bash
+# Terminal 1
+uv run crew_acp_server.py
+
+# Terminal 2
+uv run smolagents_acp_server.py
+```
+
+Run the ACP client to trigger the agent workflow:
+
+```bash
+uv run acp_client.py
+```
+
+Output:
+
+A general summary from the first agent
+
+A fact-checked and updated version from the second agent
+
+## 📬 Stay Updated with Our Newsletter!
+**Get a FREE Data Science eBook** 📖 with 150+ essential lessons in Data Science when you subscribe to our newsletter! Stay in the loop with the latest tutorials, insights, and exclusive resources. [Subscribe now!](https://join.dailydoseofds.com)
+
+[![Daily Dose of Data Science Newsletter](https://github.com/patchy631/ai-engineering/blob/main/resources/join_ddods.png)](https://join.dailydoseofds.com)
+
+---
+
+## Contribution
+
+Contributions are welcome! Please fork the repository and submit a pull request with your improvements. 

acp-code/acp_client.py

@@ -0,0 +1,24 @@
+import asyncio
+from acp_sdk.client import Client
+
+async def run_workflow() -> None:
+    async with Client(base_url="http://localhost:8000") as drafter, \
+        Client(base_url="http://localhost:8001") as verifier:
+        topic = "Impact of climate change on agriculture in 2025."
+        
+        response1 = await drafter.run_sync(
+            agent="research_drafter",
+            input=topic
+        )
+        draft = response1.output[0].parts[0].content
+        print(f"\nDraft Summary:\n{draft}")
+
+        response2 = await verifier.run_sync(
+            agent="research_verifier",
+            input=f"Enhance the following research summary using the latest information available online providing a more accurate and updated version:\n{draft}"
+        )
+        final_summary = response2.output[0].parts[0].content
+        print(f"\nVerified & Enriched Summary:\n{final_summary}")
+
+if __name__ == "__main__":
+    asyncio.run(run_workflow())

acp-code/crew_acp_server.py

@@ -0,0 +1,36 @@
+from collections.abc import AsyncGenerator
+from acp_sdk.models import Message, MessagePart 
+from acp_sdk.server import RunYield, RunYieldResume, Server
+from crewai import Crew, Task, Agent, LLM
+
+server = Server()
+
+llm = LLM(
+    model="ollama_chat/qwen2.5:14b", 
+    base_url="http://localhost:11434", 
+    max_tokens=8192
+)
+
+@server.agent()
+async def research_drafter(input: list[Message]) -> AsyncGenerator[RunYield, RunYieldResume]:
+    """Agent that creates a general research summary on a given topic."""
+
+    agent = Agent(
+        role="Research summarizer",
+        goal="Draft an informative and structured research summary based on the topic",
+        backstory="You are a researcher who summarizes complex topics for general readers.",
+        llm=llm
+    )
+    
+    task = Task(
+        description=f"Write a brief, clear summary on: {input[0].parts[0].content}",
+        expected_output="A concise paragraph summarizing the topic",
+        agent=agent
+    )
+    
+    crew = Crew(agents=[agent], tasks=[task])
+    task_output = await crew.kickoff_async()
+    yield Message(parts=[MessagePart(content=str(task_output))])
+
+if __name__ == "__main__":
+    server.run(port=8000)

acp-code/smolagents_acp_server.py

@@ -0,0 +1,30 @@
+from collections.abc import AsyncGenerator
+from acp_sdk.models import Message, MessagePart
+from acp_sdk.server import RunYield, RunYieldResume, Server
+from smolagents import CodeAgent, LiteLLMModel, DuckDuckGoSearchTool
+import warnings
+
+warnings.filterwarnings("ignore")
+
+server = Server()
+
+model = LiteLLMModel(
+    model_id="ollama_chat/qwen2.5:14b",
+    api_base="http://localhost:11434",
+    # api_key="your-api-key",
+    num_ctx=8192
+)
+
+@server.agent()
+async def research_verifier(input: list[Message]) -> AsyncGenerator[RunYield, RunYieldResume]:
+    """Agent that fact-checks and enhances a research summary using web search."""
+
+    agent = CodeAgent(tools=[DuckDuckGoSearchTool()], model=model)
+
+    prompt = input[0].parts[0].content
+    response = agent.run(prompt)
+
+    yield Message(parts=[MessagePart(content=str(response))])
+
+if __name__ == "__main__":
+    server.run(port=8001)