phidata 2.5.14

Build AI Agents with memory, knowledge and tools.

phidata

Build AI Agents with memory, knowledge, tools and reasoning

What is phidata?

Phidata is a framework for building agentic systems, use phidata to:

• Build intelligent Agents with memory, knowledge, tools and reasoning.
• Run those agents as a software application (with a database, vectordb and api).
• Monitor, evaluate and optimize your agentic system.

Install

pip install -U phidata

Agents

Web Search Agent

Let's start by building a simple agent that can search the web, create a file web_search.py

from phi.agent import Agent
from phi.model.openai import OpenAIChat
from phi.tools.duckduckgo import DuckDuckGo

web_agent = Agent(
    name="Web Agent",
    role="Search the web for information",
    model=OpenAIChat(id="gpt-4o"),
    tools=[DuckDuckGo()],
    markdown=True,
    show_tool_calls=True,
)
web_agent.print_response("Whats happening in France?", stream=True)

Install libraries, export your OPENAI_API_KEY and run the Agent:

pip install phidata openai duckduckgo-search

export OPENAI_API_KEY=sk-xxxx

python web_search.py

Finance Agent

Lets create another agent that can query financial data, create a file finance_agent.py

from phi.agent import Agent
from phi.model.openai import OpenAIChat
from phi.tools.yfinance import YFinanceTools

finance_agent = Agent(
    name="Finance Agent",
    role="Get financial data",
    model=OpenAIChat(id="gpt-4o"),
    tools=[YFinanceTools(stock_price=True, analyst_recommendations=True, company_info=True, company_news=True)],
    instructions=["Always use tables to display data"],
    markdown=True,
    show_tool_calls=True,
)
finance_agent.print_response("Share analyst recommendations for NVDA", stream=True)

Install libraries and run the Agent:

pip install yfinance

python finance_agent.py

Team of Agents

Now lets create a team of agents using the agents above, create a file agent_team.py

from phi.agent import Agent
from phi.model.openai import OpenAIChat
from phi.tools.duckduckgo import DuckDuckGo
from phi.tools.yfinance import YFinanceTools

web_agent = Agent(
    name="Web Agent",
    role="Search the web for information",
    model=OpenAIChat(id="gpt-4o"),
    tools=[DuckDuckGo()],
    markdown=True,
    show_tool_calls=True,
)

finance_agent = Agent(
    name="Finance Agent",
    role="Get financial data",
    model=OpenAIChat(id="gpt-4o"),
    tools=[YFinanceTools(stock_price=True, analyst_recommendations=True, company_info=True, company_news=True)],
    instructions=["Always use tables to display data"],
    markdown=True,
    show_tool_calls=True,
)

agent_team = Agent(
    team=[web_agent, finance_agent],
    show_tool_calls=True,
    markdown=True,
)
agent_team.print_response("Research the web for NVDA and share analyst recommendations", stream=True)

Run the Agent team:

python agent_team.py

Reasoning Agents

Reasoning helps agents work through a problem step-by-step, backtracking and correcting as needed. Let's give the reasonining agent a simple task that gpt-4o fails at. Create a file reasoning_agent.py

from phi.agent import Agent
from phi.model.openai import OpenAIChat
from phi.cli.console import console

regular_agent = Agent(model=OpenAIChat(id="gpt-4o"), markdown=True)
reasoning_agent = Agent(
    model=OpenAIChat(id="gpt-4o-2024-08-06"),
    reasoning=True,
    markdown=True,
    structured_outputs=True,
)

task = "How many 'r' are in the word 'supercalifragilisticexpialidocious'?"

console.rule("[bold green]Regular Agent[/bold green]")
regular_agent.print_response(task, stream=True)
console.rule("[bold yellow]Reasoning Agent[/bold yellow]")
reasoning_agent.print_response(task, stream=True, show_full_reasoning=True)

Run the Reasoning Agent:

python reasoning_agent.py

RAG Agent

Instead of always inserting the "context" into the prompt, the RAG Agent can search its knowledge base (vector db) for the specific information it needs to achieve its task.

This saves tokens and improves response quality. Create a file rag_agent.py

from phi.agent import Agent
from phi.model.openai import OpenAIChat
from phi.knowledge.pdf import PDFUrlKnowledgeBase
from phi.vectordb.lancedb import LanceDb, SearchType

db_uri = "tmp/lancedb"
# Create a knowledge base from a PDF
knowledge_base = PDFUrlKnowledgeBase(
    urls=["https://phi-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
    # Use LanceDB as the vector database
    vector_db=LanceDb(table_name="recipes", uri=db_uri, search_type=SearchType.vector),
)
# Load the knowledge base: Comment out after first run
knowledge_base.load(upsert=True)

agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    # Add the knowledge base to the agent
    knowledge=knowledge_base,
    show_tool_calls=True,
    markdown=True,
)
agent.print_response("How do I make chicken and galangal in coconut milk soup")

Install libraries and run the Agent:

pip install lancedb tantivy pypdf sqlalchemy pgvector 'psycopg[binary]'

python rag_agent.py

Agent UI

Phidata gives you a UI for interacting with your agents. Let's take it for a spin, create a file playground.py

Note: Phidata does not store any data, all agent data is stored locally in a sqlite database.

from phi.agent import Agent
from phi.model.openai import OpenAIChat
from phi.storage.agent.sqlite import SqlAgentStorage
from phi.tools.duckduckgo import DuckDuckGo
from phi.tools.yfinance import YFinanceTools
from phi.playground import Playground, serve_playground_app

web_agent = Agent(
    name="Web Agent",
    role="Search the web for information",
    model=OpenAIChat(id="gpt-4o"),
    tools=[DuckDuckGo()],
    storage=SqlAgentStorage(table_name="web_agent", db_file="agents.db"),
    add_history_to_messages=True,
    markdown=True,
)

finance_agent = Agent(
    name="Finance Agent",
    role="Get financial data",
    model=OpenAIChat(id="gpt-4o"),
    tools=[YFinanceTools(stock_price=True, analyst_recommendations=True, company_info=True, company_news=True)],
    instructions=["Always use tables to display data"],
    storage=SqlAgentStorage(table_name="finance_agent", db_file="agents.db"),
    add_history_to_messages=True,
    markdown=True,
)

app = Playground(agents=[finance_agent, web_agent]).get_app()

if __name__ == "__main__":
    serve_playground_app("playground:app", reload=True)

Authenticate with phidata:

phi auth

Install dependencies and run the Agent Playground:

pip install 'fastapi[standard]' sqlalchemy

python playground.py

• Open your link provided or navigate to http://phidata.app/playground
• Select your endpoint, agent and chat with your agents!

Demo Agents

The Agent Playground includes a few demo agents that you can test with. If you have recommendations for other demo agents, please let us know in our community forum.

Monitoring & Debugging

Monitoring

Phidata comes with built-in monitoring. You can set monitoring=True on any agent to track sessions or set PHI_MONITORING=true in your environment.

Note: Run phi auth first to authenticate your local account.

from phi.agent import Agent

agent = Agent(markdown=True, monitoring=True)
agent.print_response("Share a 2 sentence horror story")

Run the agent and monitor the results on phidata.app/sessions

# You can also set the environment variable
# export PHI_MONITORING=true

python agent_monitor.py

View the agent session on phidata.app/sessions

Debugging

Phidata also includes a built-in debugger that will show debug logs in the terminal. You can set debug_mode=True on any agent to track sessions or set PHI_DEBUG=true in your environment.

from phi.agent import Agent

agent = Agent(markdown=True, debug_mode=True)
agent.print_response("Share a 2 sentence horror story")

More information

• Read the docs at docs.phidata.com
• Chat with us on discord

More examples

Agent that can write and run python code

Show code

The PythonAgent can achieve tasks by writing and running python code.

• Create a file python_agent.py

from phi.agent.python import PythonAgent
from phi.model.openai import OpenAIChat
from phi.file.local.csv import CsvFile

python_agent = PythonAgent(
    model=OpenAIChat(id="gpt-4o"),
    files=[
        CsvFile(
            path="https://phidata-public.s3.amazonaws.com/demo_data/IMDB-Movie-Data.csv",
            description="Contains information about movies from IMDB.",
        )
    ],
    markdown=True,
    pip_install=True,
    show_tool_calls=True,
)

python_agent.print_response("What is the average rating of movies?")

• Install pandas and run the python_agent.py

pip install pandas

python python_agent.py

Agent that can analyze data using SQL

Show code

The DuckDbAgent can perform data analysis using SQL.

• Create a file data_analyst.py

import json
from phi.model.openai import OpenAIChat
from phi.agent.duckdb import DuckDbAgent

data_analyst = DuckDbAgent(
    model=OpenAIChat(model="gpt-4o"),
    markdown=True,
    semantic_model=json.dumps(
        {
            "tables": [
                {
                    "name": "movies",
                    "description": "Contains information about movies from IMDB.",
                    "path": "https://phidata-public.s3.amazonaws.com/demo_data/IMDB-Movie-Data.csv",
                }
            ]
        },
        indent=2,
    ),
)

data_analyst.print_response(
    "Show me a histogram of ratings. "
    "Choose an appropriate bucket size but share how you chose it. "
    "Show me the result as a pretty ascii diagram",
    stream=True,
)

• Install duckdb and run the data_analyst.py file

pip install duckdb

python data_analyst.py

Agent that can generate structured outputs

Show code

One of our favorite LLM features is generating structured data (i.e. a pydantic model) from text. Use this feature to extract features, generate movie scripts, produce fake data etc.

Let's create a Movie Agent to write a MovieScript for us.

• Create a file movie_agent.py

from typing import List
from phi.agent import Agent
from phi.model.openai import OpenAIChat
from pydantic import BaseModel, Field

class MovieScript(BaseModel):
    setting: str = Field(..., description="Provide a nice setting for a blockbuster movie.")
    ending: str = Field(..., description="Ending of the movie. If not available, provide a happy ending.")
    genre: str = Field(..., description="Genre of the movie. If not available, select action, thriller or romantic comedy.")
    name: str = Field(..., description="Give a name to this movie")
    characters: List[str] = Field(..., description="Name of characters for this movie.")
    storyline: str = Field(..., description="3 sentence storyline for the movie. Make it exciting!")

# Agent that uses JSON mode
json_mode_agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    description="You write movie scripts.",
    response_model=MovieScript,
)

# Agent that uses structured outputs
structured_output_agent = Agent(
    model=OpenAIChat(id="gpt-4o-2024-08-06"),
    description="You write movie scripts.",
    response_model=MovieScript,
    structured_outputs=True,
)

json_mode_agent.print_response("New York")
structured_output_agent.print_response("New York")

• Run the movie_agent.py file

python movie_agent.py

• The output is an object of the MovieScript class, here's how it looks:

MovieScript(
│   setting='A bustling and vibrant New York City',
│   ending='The protagonist saves the city and reconciles with their estranged family.',
│   genre='action',
│   name='City Pulse',
│   characters=['Alex Mercer', 'Nina Castillo', 'Detective Mike Johnson'],
│   storyline='In the heart of New York City, a former cop turned vigilante, Alex Mercer, teams up with a street-smart activist, Nina Castillo, to take down a corrupt political figure who threatens to destroy the city. As they navigate through the intricate web of power and deception, they uncover shocking truths that push them to the brink of their abilities. With time running out, they must race against the clock to save New York and confront their own demons.'
)

Checkout the cookbook for more examples.

Contributions

We're an open-source project and welcome contributions, please read the contributing guide for more information.

Request a feature

• If you have a feature request, please open an issue or make a pull request.
• If you have ideas on how we can improve, please create a discussion.

Telemetry

Phidata logs which model an agent used so we can prioritize features for the most popular models.

You can disable this by setting PHI_TELEMETRY=false in your environment.

⬆️ Back to Top