Workflow

cookbook/agents/basic.py

@@ -1,7 +1,7 @@
 import asyncio  # noqa
 from typing import Iterator  # noqa
 from rich.pretty import pprint  # noqa
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 from phi.tools.yfinance import YFinanceTools
 from phi.storage.agent.postgres import PgAgentStorage
@@ -16,7 +16,7 @@
     storage=PgAgentStorage(table_name="agent_sessions", db_url="postgresql+psycopg://ai:ai@localhost:5532/ai"),
 )
 
-# run1: AgentResponse = agent.run("What is the stock price of NVDA")  # type: ignore
+# run1: RunResponse = agent.run("What is the stock price of NVDA")  # type: ignore
 # pprint(run1)
 # print("------------*******************------------")
 # print(run)
@@ -27,10 +27,10 @@
 #     print(m)
 #     print("---")
 
-# run: AgentResponse = agent.run("What is the stock price of NVDA")
+# run: RunResponse = agent.run("What is the stock price of NVDA")
 # pprint(run.content)
 
-run_stream: Iterator[AgentResponse] = agent.run(
+run_stream: Iterator[RunResponse] = agent.run(
     "What is the stock price of NVDA", stream=True, stream_intermediate_steps=True
 )
 for chunk in run_stream:
@@ -40,7 +40,7 @@
 
 
 # async def main():
-#     run: AgentResponse = await agent.arun("What is the stock price of NVDA and TSLA")
+#     run: RunResponse = await agent.arun("What is the stock price of NVDA and TSLA")
 #     pprint(run)
 #     # async for chunk in await agent.arun("What is the stock price of NVDA and TSLA", stream=True):
 #     #     print(chunk.content)

cookbook/agents/image.py

@@ -1,5 +1,5 @@
 from rich.pretty import pprint  # noqa
-from phi.agent import Agent, AgentResponse
+from phi.agent import Agent, RunResponse
 from phi.model.openai import OpenAIChat
 
 agent = Agent(
@@ -8,13 +8,13 @@
     debug_mode=True,
 )
 
-# run: AgentResponse = agent.run(
+# run: RunResponse = agent.run(
 #     "What’s in this image?",
 #     images=[
 #         "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg"
 #     ],
 # )  # type: ignore
-# run: AgentResponse = agent.run(
+# run: RunResponse = agent.run(
 #     "What’s in this image?",
 #     images=[
 #         {
@@ -23,7 +23,7 @@
 #         }
 #     ],
 # )  # type: ignore
-run: AgentResponse = agent.run(
+run: RunResponse = agent.run(
     "What are in these images? Is there any difference between them?",
     images=[
         "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg",

cookbook/agents/intermediate_steps.py

@@ -1,6 +1,6 @@
 from typing import Iterator
 from rich.pretty import pprint
-from phi.agent import Agent, AgentResponse
+from phi.agent import Agent, RunResponse
 from phi.model.openai import OpenAIChat
 from phi.tools.yfinance import YFinanceTools
 
@@ -10,7 +10,7 @@
     markdown=True,
 )
 
-run_stream: Iterator[AgentResponse] = agent.run(
+run_stream: Iterator[RunResponse] = agent.run(
     "What is the stock price of NVDA", stream=True, stream_intermediate_steps=True
 )
 for chunk in run_stream:

cookbook/agents/knowledge.py

@@ -1,5 +1,5 @@
 from rich.pretty import pprint  # noqa
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 from phi.knowledge.pdf import PDFUrlKnowledgeBase
 from phi.vectordb.pgvector import PgVector, SearchType
@@ -37,5 +37,5 @@
 
 agent.print_response("How do i make Chicken and Galangal in Coconut Milk Soup")
 
-# run1: AgentResponse = agent.run("How to make Gluai Buat Chi?")  # type: ignore
+# run1: RunResponse = agent.run("How to make Gluai Buat Chi?")  # type: ignore
 # pprint(run1)

cookbook/agents/structured_output.py

@@ -9,7 +9,7 @@
 from rich.text import Text
 from pydantic import BaseModel, Field
 
-from phi.agent import Agent, AgentResponse
+from phi.agent import Agent, RunResponse
 from phi.model.openai import OpenAIChat
 
 console = Console()
@@ -80,15 +80,15 @@ def run_agents():
         # Running movie_agent_1
         display_header("Running Agent with response_model=MovieScript", panel_title="Agent 1")
         with console.status("Running Agent 1...", spinner="dots"):
-            run_movie_agent_1: AgentResponse = movie_agent_1.run("New York")
+            run_movie_agent_1: RunResponse = movie_agent_1.run("New York")
         display_content(run_movie_agent_1.content, title="Agent 1 Response")
 
         # Running movie_agent_2
         display_header(
             "Running Agent with response_model=MovieScript and structured_outputs=True", panel_title="Agent 2"
         )
         with console.status("Running Agent 2...", spinner="dots"):
-            run_movie_agent_2: AgentResponse = movie_agent_2.run("New York")
+            run_movie_agent_2: RunResponse = movie_agent_2.run("New York")
         display_content(run_movie_agent_2.content, title="Agent 2 Response")
     except Exception as e:
         console.print(f"[bold red]Error occurred while running agents: {e}[/bold red]")
@@ -99,7 +99,7 @@ async def run_agents_async():
         # Running movie_agent_1 asynchronously
         display_header("Running Agent with response_model=MovieScript (async)", panel_title="Async Agent 1")
         with console.status("Running Agent 1...", spinner="dots"):
-            async_run_movie_agent_1: AgentResponse = await movie_agent_1.arun("New York")
+            async_run_movie_agent_1: RunResponse = await movie_agent_1.arun("New York")
         display_content(async_run_movie_agent_1.content, title="Async Agent 1 Response")
 
         # Running movie_agent_2 asynchronously
@@ -108,7 +108,7 @@ async def run_agents_async():
             panel_title="Async Agent 2",
         )
         with console.status("Running Agent 2...", spinner="dots"):
-            async_run_movie_agent_2: AgentResponse = await movie_agent_2.arun("New York")
+            async_run_movie_agent_2: RunResponse = await movie_agent_2.arun("New York")
         display_content(async_run_movie_agent_2.content, title="Async Agent 2 Response")
     except Exception as e:
         console.print(f"[bold red]Error occurred while running async agents: {e}[/bold red]")

cookbook/agents/telemetry.py

@@ -1,7 +1,7 @@
 import asyncio  # noqa
 from typing import Iterator  # noqa
 from rich.pretty import pprint  # noqa
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 from phi.tools.yfinance import YFinanceTools
 from phi.storage.agent.postgres import PgAgentStorage
@@ -16,7 +16,7 @@
     storage=PgAgentStorage(table_name="agent_sessions", db_url="postgresql+psycopg://ai:ai@localhost:5532/ai"),
 )
 
-# run1: AgentResponse = agent.run("What is the stock price of NVDA")  # type: ignore
+# run1: RunResponse = agent.run("What is the stock price of NVDA")  # type: ignore
 # pprint(run1)
 # print("------------*******************------------")
 # print(run)
@@ -27,10 +27,10 @@
 #     print(m)
 #     print("---")
 
-# run: AgentResponse = agent.run("What is the stock price of NVDA")
+# run: RunResponse = agent.run("What is the stock price of NVDA")
 # pprint(run.content)
 
-# run_stream: Iterator[AgentResponse] = agent.run(
+# run_stream: Iterator[RunResponse] = agent.run(
 #     "What is the stock price of NVDA", stream=True, stream_intermediate_steps=True
 # )
 # for chunk in run_stream:
@@ -41,7 +41,7 @@
 
 async def main():
     await agent.aprint_response("What is the stock price of NVDA and TSLA")
-    # run: AgentResponse = await agent.arun("What is the stock price of NVDA and TSLA")
+    # run: RunResponse = await agent.arun("What is the stock price of NVDA and TSLA")
     # pprint(run)
     # async for chunk in await agent.arun("What is the stock price of NVDA and TSLA", stream=True):
     #     print(chunk.content)

cookbook/providers/openai/agent.py

@@ -1,4 +1,4 @@
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 from phi.tools.yfinance import YFinanceTools
 
@@ -10,7 +10,7 @@
 )
 
 # Get the response in a variable
-# run: AgentResponse = agent.run("What is the stock price of NVDA and TSLA")
+# run: RunResponse = agent.run("What is the stock price of NVDA and TSLA")
 # print(run.content)
 
 # Print the response on the terminal

cookbook/providers/openai/agent_stream.py

@@ -1,5 +1,5 @@
 from typing import Iterator  # noqa
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 from phi.tools.yfinance import YFinanceTools
 
@@ -11,7 +11,7 @@
 )
 
 # Get the response in a variable
-# run_response: Iterator[AgentResponse] = agent.run("What is the stock price of NVDA and TSLA", stream=True)
+# run_response: Iterator[RunResponse] = agent.run("What is the stock price of NVDA and TSLA", stream=True)
 # for chunk in run_response:
 #     print(chunk.content)
 

cookbook/providers/openai/basic.py

@@ -1,10 +1,10 @@
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 
 agent = Agent(model=OpenAIChat(id="gpt-4o"), instructions=["Respond in a southern tone"], markdown=True)
 
 # Get the response in a variable
-# run: AgentResponse = agent.run("Explain simulation theory")
+# run: RunResponse = agent.run("Explain simulation theory")
 # print(run.content)
 
 # Print the response on the terminal

cookbook/providers/openai/basic_stream.py

@@ -1,11 +1,11 @@
 from typing import Iterator  # noqa
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 
 agent = Agent(model=OpenAIChat(id="gpt-4o"), instructions=["Respond in a southern tone"], markdown=True)
 
 # Get the response in a variable
-# run_response: Iterator[AgentResponse] = agent.run("Explain simulation theory", stream=True)
+# run_response: Iterator[RunResponse] = agent.run("Explain simulation theory", stream=True)
 # for chunk in run_response:
 #     print(chunk.content)
 

cookbook/providers/openai/structured_output.py

@@ -1,7 +1,7 @@
 from typing import List
 from rich.pretty import pprint  # noqa
 from pydantic import BaseModel, Field
-from phi.agent import Agent, AgentResponse  # noqa
+from phi.agent import Agent, RunResponse  # noqa
 from phi.model.openai import OpenAIChat
 
 
@@ -25,7 +25,7 @@ class MovieScript(BaseModel):
 )
 
 # Get the response in a variable
-# run: AgentResponse = movie_agent.run("New York")
+# run: RunResponse = movie_agent.run("New York")
 # pprint(run.content)
 
 movie_writer.print_response("New York")

cookbook/workflows/news_article.py

@@ -1,11 +1,13 @@
+from textwrap import dedent
 from typing import Optional, Iterator
 
 from pydantic import BaseModel, Field
 
-from phi.agent import Agent, AgentResponse
+from phi.agent import Agent, RunResponse
 from phi.workflow import Workflow
 from phi.tools.duckduckgo import DuckDuckGo
-from phi.utils.pprint import pprint_agent_response_stream
+from phi.tools.newspaper4k import Newspaper4k
+from phi.utils.pprint import pprint_run_response
 from phi.utils.log import logger
 
 
@@ -27,22 +29,63 @@ class WriteNewsReport(Workflow):
         response_model=NewsArticles,
     )
 
-    writer: Agent = Agent()
+    writer: Agent = Agent(
+        name="Writer",
+        tools=[Newspaper4k()],
+        description="You are a Senior NYT Editor and your task is to write a NYT cover story due tomorrow.",
+        instructions=[
+            "You will be provided with news articles and their links.",
+            "Carefully read each article and think about the contents",
+            "Then generate a final New York Times worthy article in the <article_format> provided below.",
+            "Break the article into sections and provide key takeaways at the end.",
+            "Make sure the title is catchy and engaging.",
+            "Give the section relevant titles and provide details/facts/processes in each section."
+            "Ignore articles that you cannot read or understand.",
+            "REMEMBER: you are writing for the New York Times, so the quality of the article is important.",
+        ],
+        expected_output=dedent("""\
+        An engaging, informative, and well-structured article in the following format:
+        <article_format>
+        ## Engaging Article Title
 
-    def run(self, topic: str) -> Iterator[AgentResponse]:
+        ### Overview
+        {give a brief introduction of the article and why the user should read this report}
+        {make this section engaging and create a hook for the reader}
+
+        ### Section 1
+        {break the article into sections}
+        {provide details/facts/processes in this section}
+
+        ... more sections as necessary...
+
+        ### Takeaways
+        {provide key takeaways from the article}
+
+        ### References
+        - [Title](url)
+        - [Title](url)
+        - [Title](url)
+        </article_format>
+        """),
+    )
+
+    def run(self, topic: str) -> Iterator[RunResponse]:
         logger.info(f"Researching articles on: {topic}")
-        # research: AgentResponse = self.researcher.run(topic)
-        # if research.content and isinstance(research.content, NewsArticles) and research.content.articles:
-        #     logger.info(f"Research identified {len(research.content.articles)} articles.")
-        # else:
-        #     logger.error("No articles found.")
-        #     return
+        research: RunResponse = self.researcher.run(topic)
+        if research and research.content and isinstance(research.content, NewsArticles) and research.content.articles:
+            logger.info(f"Researcher identified {len(research.content.articles)} articles.")
+        else:
+            yield RunResponse(
+                run_id=self.run_id,
+                content=f"Sorry could not find any articles on the topic: {topic}",
+            )
+            return
 
         logger.info("Reading each article and writing a report.")
-        # writer.print_response(research.content.model_dump_json(indent=2), markdown=True, show_message=False)
-        yield from self.writer.run(f"write 1 sentence on {topic}", stream=True)
+        yield from self.writer.run(research.content.model_dump_json(indent=2), stream=True)
 
 
 # Run workflow
-story = WriteNewsReport(debug_mode=False).run(topic="avocado toast")
-pprint_agent_response_stream(story, markdown=True, show_time=True)
+story = WriteNewsReport(debug_mode=False).run(topic="IBM Hashicorp Acquisition")
+# Print the response
+pprint_run_response(story, markdown=True, show_time=True)