### Install and start MLflow server

Source: https://mlflow.org/docs/latest/genai/eval-monitor/scorers/llm-judge/custom-judges/create-custom-judge

Install MLflow and start the server to use the UI Judge Builder. Run these commands in a terminal.

```bash
pip install mlflow

mlflow server
```

--------------------------------

### Install MLflow GenAI extra and start server (uv)

Source: https://mlflow.org/docs/latest/genai/tracing/token-usage-cost

Install the MLflow GenAI extra and start the tracking server using uv. Required for cost tracking with MLflow version 3.10.0 or later.

```bash
uv add 'mlflow[genai]>=3.10.0'

uv run mlflow server
```

--------------------------------

### Install dependencies for Agno tracing example

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/agno

Install the required packages: mlflow>=3.3, agno, anthropic, and yfinance.

```bash
pip install 'mlflow>=3.3' agno anthropic yfinance
```

--------------------------------

### Run the interactive setup command for MLflow tracing

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/qwen_code

Run `mlflow-qwen-code setup` to start an interactive flow that configures the tracing hook. The output shows the prompts and resulting configuration files. Use `--non-interactive` or `-y` to skip the interactive flow and use defaults.

```text
~/my-project $ mlflow-qwen-code setup



Configure MLflow tracing for Qwen Code

? Where should MLflow tracing be installed?

  ● Project  ./.qwen/  (default)

  ○ User  ~/.qwen/



  ↑/↓ to move, enter to select



✓ Registered Stop hook in /path/to/project/.qwen/settings.json



  MLflow tracking URI [http://localhost:5000] http://localhost:5000

  MLflow experiment ID [0] 1



✓ Wrote tracing config to /path/to/project/.qwen/mlflow-tracing.json



Next steps

  1. Start the MLflow tracking server in a separate terminal:

       mlflow server --port 5000

  2. Launch qwen — traces appear at http://localhost:5000 after each turn.
```

--------------------------------

### Install MLflow GenAI extra and start server (pip)

Source: https://mlflow.org/docs/latest/genai/tracing/token-usage-cost

Install the MLflow GenAI extra and start the tracking server using pip. Required for cost tracking with MLflow version 3.10.0 or later.

```bash
pip install 'mlflow[genai]>=3.10.0'

mlflow server
```

--------------------------------

### Install OpenAI SDK

Source: https://mlflow.org/docs/latest/genai/eval-monitor/quickstart

Install the OpenAI Python package, required for the code examples in this guide. MLflow's evaluation framework works with any LLM provider, but these examples use the OpenAI SDK.

```bash
pip install openai
```

--------------------------------

### Install MLflow with pip

Source: https://mlflow.org/docs/latest/genai/eval-monitor/scorers/custom/tutorial

Install MLflow and required packages before running the examples. Run this in a bash shell.

```bash
pip install --upgrade mlflow
```

--------------------------------

### Install MLflow and Start the Server

Source: https://mlflow.org/docs/latest/genai/governance/ai-gateway/quickstart

Installs MLflow with the genai extra and starts the tracking server on port 5000. The AI Gateway is built into the MLflow Tracking Server and will be ready at http://localhost:5000. Requires a SQL-based backend store and the FastAPI tracking server; the default SQLite and FastAPI configuration is sufficient.

```bash
pip install 'mlflow[genai]'

mlflow server --port 5000
```

--------------------------------

### Install MLflow and LangChain OpenAI

Source: https://mlflow.org/docs/latest/genai/version-tracking/quickstart

Install MLflow 3.0 or higher and the LangChain OpenAI integration package. Required before using the quickstart's LLM and AI agent features.

```bash
pip install --upgrade mlflow

pip install langchain-openai
```

--------------------------------

### Start MLflow Server with uv

Source: https://mlflow.org/docs/latest/genai/eval-monitor/quickstart

Start a local MLflow Tracking Server using the uvx command, which requires installing the uv package manager first. This is one of three methods to start the server.

```shell
uvx mlflow server
```

--------------------------------

### Install the MLflow Qwen Code tracing plugin

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/qwen_code

Install the @mlflow/qwen-code npm package globally to enable MLflow tracing for Qwen Code. Alternatively, use npx to run the setup command without global installation.

```bash
npm install -g @mlflow/qwen-code
```

--------------------------------

### Basic AG2 tracing example with autolog

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/ag2

Basic example that imports os and typing utilities before enabling AG2 autologging. Use this as a starting point for instrumenting AG2 agent execution with MLflow tracing.

```python
import os

from typing import Annotated, Literal
```

--------------------------------

### Install Omnigent and MLflow

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/omnigent

Install the required packages using uv. This is the first step in setting up tracing.

```bash
uv pip install omnigent mlflow
```

--------------------------------

### Quickstart: Log OpenAI traces to a remote MLflow server

Source: https://mlflow.org/docs/latest/genai/tracing/lightweight-sdk

Sets the tracking URI and experiment for the Production Tracing SDK, then imports openai for tracing LLM applications. Use this when logging traces to an experiment on a remote MLflow server.

```python
import mlflow

import openai



# Set the tracking URI to your MLflow server

mlflow.set_tracking_uri("http://your-mlflow-server:5000")

mlflow.set_experiment("genai-production-monitoring")
```

--------------------------------

### Manual Installation of n8n-nodes-mlflow

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/n8n

Install the community node manually within your n8n directory and start the service.

```bash
cd ~/.n8n

npm install n8n-nodes-mlflow

n8n start
```

--------------------------------

### Setting Up Working Environment and FAISS Database

Source: https://mlflow.org/docs/latest/genai/flavors/langchain/notebooks/langchain-retriever

Sets up a temporary directory, defines paths for documents and a FAISS index, fetches documents from the given URLs, and creates a FAISS database with default chunk settings.

```python
temporary_directory = tempfile.mkdtemp()

doc_path = os.path.join(temporary_directory, "docs.txt")
persist_dir = os.path.join(temporary_directory, "faiss_index")

url_listings = [
  "https://www.archives.gov/milestone-documents/act-establishing-yellowstone-national-park#transcript",
  "https://www.archives.gov/milestone-documents/sherman-anti-trust-act#transcript",
]

fetch_and_save_documents(url_listings, doc_path)

vector_db = create_faiss_database(doc_path, persist_dir)
```

--------------------------------

### Set Up OTLP Exporter for MLflow Traces

Source: https://mlflow.org/docs/latest/genai/tracing/opentelemetry/export

Install the OTLP exporter package and configure the environment variables to export MLflow traces to an OpenTelemetry Collector. Set OTEL_EXPORTER_OTLP_TRACES_ENDPOINT before starting any trace, and optionally set OTEL_SERVICE_NAME to group traces. The Python example creates a span and exports it to the collector.

```bash
pip install opentelemetry-exporter-otlp
```

```python
import mlflow

import os



# Set the endpoint of the OpenTelemetry Collector

os.environ["OTEL_EXPORTER_OTLP_TRACES_ENDPOINT"] = "http://localhost:4317/v1/traces"

# Optionally, set the service name to group traces

os.environ["OTEL_SERVICE_NAME"] = "your-service-name"



# Trace will be exported to the OTel collector

with mlflow.start_span(name="foo") as span:

    span.set_inputs({"a": 1})

    span.set_outputs({"b": 2})
```

--------------------------------

### Create Index and Run Query, Chat, and Retriever Examples

Source: https://mlflow.org/docs/latest/genai/flavors/llama-index/notebooks/llama_index_quickstart

Creates a VectorStoreIndex from an example document and demonstrates three types of inference: a query engine, a chat engine with a system message, and a retriever. Each result is printed to the console.

```python
print("------------- Example Document used to Enrich LLM Context -------------")

llama_index_example_document = Document.example()

print(llama_index_example_document)



index = VectorStoreIndex.from_documents([llama_index_example_document])



print("

------------- Example Query Engine -------------")

query_response = index.as_query_engine().query("What is llama_index?")

print(query_response)



print("

------------- Example Chat Engine  -------------")

chat_response = index.as_chat_engine().chat(

  "What is llama_index?",

  chat_history=[ChatMessage(role="system", content="You are an expert on RAG!")],

)

print(chat_response)





print("

------------- Example Retriever   -------------")

retriever_response = index.as_retriever().retrieve("What is llama_index?")

print(retriever_response)
```

--------------------------------

### Update MLflow and install OpenAI

Source: https://mlflow.org/docs/latest/genai/eval-monitor/scorers/custom/code-examples

Install the latest MLflow and OpenAI packages using pip. Required for the example app that uses the OpenAI client.

```bash
pip install --upgrade mlflow openai
```

--------------------------------

### Setup and Generate Traces with Initial Judge

Source: https://mlflow.org/docs/latest/genai/eval-monitor/scorers/llm-judge/alignment

Creates an experiment and an initial judge named 'product_quality' using make_judge, then generates 15 traces and runs the judge on them to produce initial assessments. The judge uses field-based evaluation with inputs and outputs extracted from the trace spans. Requires at least 10 traces for alignment.

```python
from mlflow.genai.judges import make_judge

from mlflow.genai.judges.optimizers import MemAlignOptimizer

from mlflow.entities import AssessmentSource, AssessmentSourceType

from typing import Literal

import mlflow



# Create experiment and initial judge

experiment_id = mlflow.create_experiment("product-quality-alignment")

mlflow.set_experiment(experiment_id=experiment_id)



initial_judge = make_judge(

    name="product_quality",

    instructions=(

        "Evaluate if the product description in {{ outputs }} "

        "is accurate and helpful for the query in {{ inputs }}. "

        "Rate as: excellent, good, fair, or poor"

    ),

    feedback_value_type=Literal["excellent", "good", "fair", "poor"],

    model="anthropic:/claude-opus-4-1-20250805",

)



# Generate traces from your application (minimum 10 required)

traces = []

for i in range(15):  # Generate 15 traces (more than minimum of 10)

    with mlflow.start_span(f"product_description_{i}") as span:

        # Your application logic

        query = f"Product query {i}"

        description = f"Product description for query {i}"

        span.set_inputs({"query": query})

        span.set_outputs({"description": description})

        traces.append(span.trace_id)



# Run the judge on these traces to get initial assessments

for trace_id in traces:

    trace = mlflow.get_trace(trace_id)



    # Extract inputs and outputs from the trace for field-based evaluation

    inputs = trace.data.spans[0].inputs  # Get inputs from trace

    outputs = trace.data.spans[0].outputs  # Get outputs from trace



    # Judge evaluates using field-based approach (inputs/outputs)

    judge_result = initial_judge(inputs=inputs, outputs=outputs)

    # Judge's assessment is automatically logged when called
```

--------------------------------

### Start MLflow Server with Docker Compose

Source: https://mlflow.org/docs/latest/genai/eval-monitor/quickstart

Start a local MLflow server with PostgreSQL and MinIO using the provided Docker Compose file. Includes cloning the MLflow repo, setting up the docker-compose directory, and copying the environment file.

```shell
git clone --depth 1 --filter=blob:none --sparse https://github.com/mlflow/mlflow.git

cd mlflow

git sparse-checkout set docker-compose

cd docker-compose

cp .env.dev.example .env

docker compose up -d
```

--------------------------------

### Start MLflow Server with pip

Source: https://mlflow.org/docs/latest/genai/eval-monitor/quickstart

Install the mlflow Python package via pip and start a local MLflow server. Requires Python 3.10+.

```shell
pip install --upgrade mlflow

mlflow server
```

--------------------------------

### Complete example: optimize Pydantic AI customer support agent

Source: https://mlflow.org/docs/latest/genai/prompt-registry/optimize-prompts/pydantic-ai-optimization

This example shows the full workflow: register initial prompts, define a prediction function using Pydantic AI, prepare training data, optimize with GepaPromptOptimizer and Correctness scorer, and use the optimized prompts. The agent automatically uses the optimized prompts via @latest.

```python
import mlflow

from mlflow.genai.scorers import Correctness

from mlflow.genai.optimize.optimizers import GepaPromptOptimizer

from pydantic_ai import Agent



# If you're inside notebooks, please uncomment the following lines.

# import nest_asyncio

# nest_asyncio.apply()



# Step 1: Register your initial prompts

system_prompt = mlflow.genai.register_prompt(

    name="customer-support-system",

    template="You are a helpful customer support agent for an e-commerce platform. "

    "Assist customers with their questions about orders, returns, and products.",

)



user_prompt = mlflow.genai.register_prompt(

    name="customer-support-query",

    template="Customer inquiry: {{query}}",

)





# Step 2: Create a prediction function that uses Pydantic AI

@mlflow.trace

def predict_fn(query):

    # Load prompts from registry

    system_prompt = mlflow.genai.load_prompt("prompts:/customer-support-system@latest")

    user_prompt = mlflow.genai.load_prompt("prompts:/customer-support-query@latest")



    # Initialize agent with system prompt

    agent = Agent(

        model="openai:gpt-5-mini",

        system_prompt=system_prompt.template,

    )



    # Format user message and run agent

    formatted_query = user_prompt.format(query=query)

    result = agent.run_sync(formatted_query)



    return result.output





# Step 3: Prepare training data

dataset = [

    {

        "inputs": {"query": "Where is my order #12345?"},

        "expectations": {

            "expected_response": "I'd be happy to help you track your order #12345. "

            "Please check your email for a tracking link, or I can look it up for you if you provide your email address."

        },

    },

    {

        "inputs": {"query": "How do I return a defective product?"},

        "expectations": {

            "expected_response": "I'm sorry to hear your product is defective. You can initiate a return "

            "through your account's order history within 30 days of purchase. We'll send you a prepaid shipping label."

        },

    },

    {

        "inputs": {"query": "Do you have this item in blue?"},

        "expectations": {

            "expected_response": "I'd be happy to check product availability for you. "

            "Could you please provide the product name or SKU so I can verify if it's available in blue?"

        },

    },

    # more data...

]



# Step 4: Optimize the prompts

result = mlflow.genai.optimize_prompts(

    predict_fn=predict_fn,

    train_data=dataset,

    prompt_uris=[system_prompt.uri, user_prompt.uri],

    optimizer=GepaPromptOptimizer(reflection_model="openai:/gpt-5"),

    scorers=[Correctness(model="openai:/gpt-5")],

)



# Step 5: Use the optimized prompts

optimized_system_prompt = result.optimized_prompts[0]

optimized_user_prompt = result.optimized_prompts[1]



print(f"Optimized system prompt URI: {optimized_system_prompt.uri}")

print(f"Optimized system template: {optimized_system_prompt.template}")

print(f"Optimized user prompt URI: {optimized_user_prompt.uri}")

print(f"Optimized user template: {optimized_user_prompt.template}")



# Since your agent already uses @latest, it will automatically use the optimized prompts

predict_fn("Can I get a refund for order #67890?")
```

--------------------------------

### Start AgentServer in start_server.py

Source: https://mlflow.org/docs/latest/genai/serving/agent-server

Create a start_server.py file that imports the agent module to register functions, creates an AgentServer for the 'ResponsesAgent' agent, and optionally sets up MLflow git-based version tracking. The main function runs the server with an import string for multi-worker support.

```python
# Need to import the agent to register the functions with the server

import agent  # noqa: F401

from mlflow.genai.agent_server import (

    AgentServer,

    setup_mlflow_git_based_version_tracking,

)



agent_server = AgentServer("ResponsesAgent")

app = agent_server.app



# Optionally, set up MLflow git-based version tracking

# to correspond your agent's traces to a specific git commit

setup_mlflow_git_based_version_tracking()





def main():

    # To support multiple workers, pass the app as an import string

    agent_server.run(app_import_string="start_server:app")



if __name__ == "__main__":

    main()

```

--------------------------------

### Install the MLflow Codex Tracing Plugin

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/codex

Installs the @mlflow/codex npm package globally so MLflow can automatically trace Codex CLI conversations. Requires MLflow tracking server, Codex CLI, and Node.js 18+. Alternatively, invoke the bin through npx (e.g., npx @mlflow/codex setup) instead of installing globally.

```bash
npm install -g @mlflow/codex
```

--------------------------------

### Start Open WebUI with open-webui serve

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/open-webui

Run this command to start the Open WebUI server, which will host the chat interface that the MLflow pipeline intercepts.

```bash
open-webui serve
```

--------------------------------

### Install Python dependencies for LangChain tracing (shell)

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/langchain

Confirmed working versions for the Python example: openai, langchain, langchain-openai, mlflow, and tiktoken.

```shell
pip install openai==3.3.1 langchain==1.3.17 langchain-openai==1.6.0 mlflow==3.15.1 tiktoken==0.14.0
```

--------------------------------

### Start MLflow server with local budget tracker

Source: https://mlflow.org/docs/latest/genai/governance/ai-gateway/budget-alerts-limits

Command to start the MLflow server using the local budget tracker strategy. No extra environment variables are required.

```bash
# No extra environment variables required
mlflow server --host 0.0.0.0 --port 5000
```

--------------------------------

### Define per-row guidelines for document extraction with ExpectationsGuidelines

Source: https://mlflow.org/docs/latest/genai/eval-monitor/scorers/llm-judge/guidelines

Starts a per-row guidelines example for a document extraction app, importing ExpectationsGuidelines and defining a dataset with document-type specific inputs. The snippet is truncated in the source, showing only the beginning of the document_extraction_data list with an invoice example.

```python
from mlflow.genai.scorers import ExpectationsGuidelines

import mlflow





# Dataset with document-type specific guidelines

document_extraction_data = [

    {

        "inputs": {

            "document_type": "invoice",

            "document_text": "Invoice #INV-2024-001\nBill To: Acme Corp\nAmount: $1,234.56\nDue Date: 2024-03-15",

        },
```

--------------------------------

### Run MLflow demo locally with uvx mlflow demo

Source: https://mlflow.org/docs/latest/genai/demo

Launches a local MLflow server pre-loaded with the same sample data as the live demo, providing full read-write access for experimentation and customization.

```bash
uvx mlflow demo
```

--------------------------------

### Start MLflow Tracking Server with SQLite backend

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/google-adk

Start the MLflow Tracking Server with a SQL-based backend store. This example uses SQLite; for other SQL databases (PostgreSQL, MySQL, MSSQL) change the store URI. OpenTelemetry ingestion is not supported with file-based backend stores.

```bash
mlflow server --backend-store-uri sqlite:///mlflow.db --port 5000
```

--------------------------------

### Set MLFLOW_GATEWAY_BUDGET_REFRESH_INTERVAL and start server

Source: https://mlflow.org/docs/latest/genai/governance/ai-gateway/budget-alerts-limits

Use this bash example to override the default 600-second budget policy refresh interval. Set the environment variable to a lower value to pick up policy changes faster or a higher value to reduce database load, then start the MLflow server.

```bash
export MLFLOW_GATEWAY_BUDGET_REFRESH_INTERVAL=30

mlflow server --host 0.0.0.0 --port 5000
```

--------------------------------

### Conversation Simulation Quick Start

Source: https://mlflow.org/docs/latest/genai/eval-monitor/running-evaluation/conversation-simulation

Complete example that simulates conversations and evaluates them using MLflow GenAI. Sets up test cases with goals, personas, and simulation guidelines, then creates a ConversationSimulator.

```python
import os

import mlflow

from mlflow.genai.simulators import ConversationSimulator

from mlflow.genai.scorers import ConversationCompleteness, Safety

from openai import OpenAI



os.environ["OPENAI_API_KEY"] = "your-api-key-here"  # Replace with your API key



client = OpenAI()



# 1. Define test cases with goals (required) and optional persona/context

test_cases = [

    {

        "goal": "Learn how to track experiments in MLflow",

    },

    {

        "goal": "Debug a model deployment issue",

        "persona": "You are a frustrated data scientist who has been stuck on this issue for hours",

    },

    {

        
```

--------------------------------

### TripCrew class with Crew setup

Source: https://mlflow.org/docs/latest/genai/tracing/integrations/listing/crewai

Defines the TripCrew class that initializes origin, cities, date range, and interests, then runs a Crew with two agents and two tasks. The crew uses verbose output, memory, and knowledge features. This snippet is part of a larger trip planning example that includes task definitions and agent setup.

```python
class TripCrew:

    def __init__(self, origin, cities, date_range, interests):
        self.cities = cities
        self.origin = origin
        self.interests = interests
        self.date_range = date_range

    def run(self):
        agents = TripAgents()
        tasks = TripTasks()

        city_selector_agent = agents.city_selection_agent()
        local_expert_agent = agents.local_expert()

        identify_task = tasks.identify_task(
            city_selector_agent,
            self.origin,
            self.cities,
            self.interests,
            self.date_range,
        )
        gather_task = tasks.gather_task(
            local_expert_agent, self.origin, self.interests, self.date_range
        )

        crew = Crew(
            agents=[city_selector_agent, local_expert_agent],
            tasks=[identify_task, gather_task],
            verbose=True,
            memory=True,
            knowledge={...},  # knowledge config here (if any) or omitted for now; see docs for details on enabling knowledge for your model. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#knowledge-parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/concepts/knowledge#parameters. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crewai/docs/build/crewai/agents/01-overview-agent.md. For more info, see https://docs.crewai.com/content/crew
```