SQLite - PraisonAI

Overview

SQLite tool allows you to query and manage SQLite databases. No server required - perfect for local development and embedded applications. The user asks a data question; the agent runs SQL against the database and returns rows.

Installation

pip install "praisonai[tools]"

No additional dependencies required - SQLite is built into Python!

Quick Start

Simple Usage

from praisonai_tools import SQLiteTool

# Initialize
sqlite = SQLiteTool(path="mydata.db")

# Query
results = sqlite.query("SELECT * FROM users LIMIT 5")
print(results)

With Configuration

Use the same tool with an agent — see Usage with Agent below, or pass env vars and options from the sections above.

Usage with Agent

from praisonaiagents import Agent
from praisonai_tools import SQLiteTool

sqlite = SQLiteTool(path="mydata.db")

agent = Agent(
    name="DBAnalyst",
    instructions="You are a database analyst. Use SQLite to query data.",
    tools=[sqlite]
)

response = agent.chat("Show me all tables in the database")
print(response)

Available Methods

query(sql)

Execute a SQL query.

from praisonai_tools import SQLiteTool

sqlite = SQLiteTool(path="mydata.db")
results = sqlite.query("SELECT * FROM users WHERE active = 1")

execute(sql)

Execute a SQL statement (INSERT, UPDATE, DELETE, CREATE).

sqlite.execute("CREATE TABLE IF NOT EXISTS users (id INTEGER PRIMARY KEY, name TEXT)")
sqlite.execute("INSERT INTO users (name) VALUES ('Alice')")

list_tables()

List all tables in the database.

tables = sqlite.list_tables()

Configuration Options

sqlite = SQLiteTool(
    path="mydata.db",      # Database file path
    timeout=30             # Connection timeout
)

Function-Based Usage

from praisonai_tools import query_sqlite

# Quick query
results = query_sqlite("SELECT * FROM users", path="mydata.db")

In-Memory Database

# Use :memory: for temporary in-memory database
sqlite = SQLiteTool(path=":memory:")
sqlite.execute("CREATE TABLE test (id INTEGER, value TEXT)")

Error Handling

from praisonai_tools import SQLiteTool

sqlite = SQLiteTool(path="mydata.db")
result = sqlite.query("SELECT * FROM users")

if isinstance(result, dict) and "error" in result:
    print(f"Error: {result['error']}")
else:
    for row in result:
        print(row)

Common Errors

Error	Cause	Solution
`no such table`	Table doesn’t exist	Create table first
`database is locked`	Concurrent access	Close other connections
`disk I/O error`	File permission issue	Check file permissions

How It Works

Best Practices

Inspect the schema first

List tables and columns before querying so the agent writes valid SQL.

Use parameterised queries

Bind values instead of string-formatting them to avoid SQL injection.

Limit result size

Add LIMIT so the agent works with a focused result set.

PostgreSQL

PostgreSQL database

MySQL

MySQL database

DuckDB

Analytics database

​Overview

​Installation

​Quick Start

​Usage with Agent

​Available Methods

​query(sql)

​execute(sql)

​list_tables()

​Configuration Options

​Function-Based Usage

​In-Memory Database

​Error Handling

​Common Errors

​How It Works

​Best Practices

​Related Tools