> ## Documentation Index
> Fetch the complete documentation index at: https://docs.praison.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Data Analyst Agent

> Learn how to create AI agents for data analysis and insights generation.

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
flowchart LR
    In[Data File] --> Agent[Data Analyst]
    Agent --> Out[Insights Report]
    
    style In fill:#8B0000,color:#fff
    style Agent fill:#2E8B57,color:#fff
    style Out fill:#8B0000,color:#fff
```

Data analysis agent with CSV/Excel tools for reading, analyzing, and exporting data.

***

## Simple

**Agents: 1** — Single agent with data tools handles file operations and analysis.

### Workflow

1. Read data from CSV/Excel
2. Analyze with filtering, grouping
3. Generate statistical summaries

### Setup

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
pip install praisonaiagents praisonai pandas openpyxl
export OPENAI_API_KEY="your-key"
```

### Run — Python

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import Agent
from praisonaiagents import read_csv, get_summary, filter_data

agent = Agent(
    name="DataAnalyst",
    instructions="You are a data analyst. Analyze data and provide insights.",
    tools=[read_csv, get_summary, filter_data]
)

result = agent.start("Read sales_data.csv and provide a summary")
print(result)
```

### Run — CLI

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai "Analyze data.csv and summarize key metrics" --tools pandas
```

### Run — agents.yaml

```yaml theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
framework: praisonai
topic: Data Analysis
roles:
  data_analyst:
    role: Data Analyst
    goal: Analyze data and generate insights
    backstory: You are an expert data analyst
    tools:
      - read_csv
      - get_summary
      - filter_data
    tasks:
      analyze_data:
        description: Read sales_data.csv and provide a summary
        expected_output: A data analysis report
```

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai agents.yaml
```

### Serve API

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import Agent
from praisonaiagents import read_csv, get_summary, filter_data

agent = Agent(
    name="DataAnalyst",
    instructions="You are a data analyst.",
    tools=[read_csv, get_summary, filter_data]
)

agent.launch(port=8080)
```

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
curl -X POST http://localhost:8080/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Summarize the uploaded data"}'
```

***

## Advanced Workflow (All Features)

**Agents: 1** — Single agent with memory, persistence, structured output, and session resumability.

### Workflow

1. Initialize session for analysis tracking
2. Configure SQLite persistence for analysis history
3. Read and analyze data with structured output
4. Store insights in memory for comparison
5. Resume session for iterative analysis

### Setup

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
pip install praisonaiagents praisonai pandas openpyxl pydantic
export OPENAI_API_KEY="your-key"
```

### Run — Python

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import Agent, Task, AgentTeam, Session
from praisonaiagents import read_csv, get_summary, filter_data
from pydantic import BaseModel

# Structured output schema
class DataInsights(BaseModel):
    dataset: str
    row_count: int
    key_metrics: list[str]
    trends: list[str]
    recommendations: list[str]

# Create session for analysis tracking
session = Session(session_id="analysis-001", user_id="user-1")

# Agent with memory and tools
agent = Agent(
    name="DataAnalyst",
    instructions="Analyze data and return structured insights.",
    tools=[read_csv, get_summary, filter_data],
    memory=True
)

# Task with structured output
task = Task(
    description="Read sales_data.csv and provide structured insights",
    expected_output="Structured data analysis",
    agent=agent,
    output_pydantic=DataInsights
)

# Run with SQLite persistence
agents = AgentTeam(
    agents=[agent],
    tasks=[task],
    memory=True
)

result = agents.start()
print(result)

# Resume later
session2 = Session(session_id="analysis-001", user_id="user-1")
history = session2.search_memory("sales")
```

### Run — CLI

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai "Analyze data.csv" --tools pandas --memory --verbose
```

### Run — agents.yaml

```yaml theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
framework: praisonai
topic: Data Analysis
memory: true
memory_config:
  provider: sqlite
  db_path: analysis.db
roles:
  data_analyst:
    role: Data Analyst
    goal: Analyze data with structured output
    backstory: You are an expert data analyst
    tools:
      - read_csv
      - get_summary
      - filter_data
    memory: true
    tasks:
      analyze_data:
        description: Read sales_data.csv and provide structured insights
        expected_output: Structured data analysis
        output_json:
          dataset: string
          row_count: number
          key_metrics: array
          trends: array
          recommendations: array
```

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai agents.yaml --verbose
```

### Serve API

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import Agent
from praisonaiagents import read_csv, get_summary, filter_data

agent = Agent(
    name="DataAnalyst",
    instructions="Analyze data and return structured insights.",
    tools=[read_csv, get_summary, filter_data],
    memory=True
)

agent.launch(port=8080)
```

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
curl -X POST http://localhost:8080/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Analyze data", "session_id": "analysis-001"}'
```

***

## Monitor / Verify

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai "test analysis" --tools pandas --verbose
```

## Cleanup

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
rm -f analysis.db
```

## Features Demonstrated

| Feature           | Implementation                 |
| ----------------- | ------------------------------ |
| Workflow          | Multi-tool data analysis       |
| DB Persistence    | SQLite via `memory_config`     |
| Observability     | `--verbose` flag               |
| Tools             | pandas (read, summary, filter) |
| Resumability      | `Session` with `session_id`    |
| Structured Output | Pydantic `DataInsights` model  |

## Next Steps

* [Finance Agent](/agents/finance) for stock analysis
* [Research Agent](/agents/research) for web research
* [Memory](/features/advanced-memory) for persistent context