SQLTools enable an Agent to run SQL queries and interact with databases.

​
Prerequisites

The following example requires the sqlalchemy library and a database URL.

pip install -U sqlalchemy

You will also need a database. The following example uses a Postgres database running in a Docker container.

 docker run -d \
  -e POSTGRES_DB=ai \
  -e POSTGRES_USER=ai \
  -e POSTGRES_PASSWORD=ai \
  -e PGDATA=/var/lib/postgresql/data/pgdata \
  -v pgvolume:/var/lib/postgresql/data \
  -p 5532:5432 \
  --name pgvector \
  phidata/pgvector:16

​
Example

The following agent will run a SQL query to list all tables in the database and describe the contents of one of the tables.

cookbook/tools/sql_tools.py
from phi.agent import Agent
from phi.tools.sql import SQLTools

db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"

agent = Agent(tools=[SQLTools(db_url=db_url)])
agent.print_response("List the tables in the database. Tell me about contents of one of the tables", markdown=True)

​
Toolkit Params

ParameterTypeDefaultDescription
db_urlstr-The URL for connecting to the database.
db_engineEngine-The database engine used for connections and operations.
userstr-The username for database authentication.
passwordstr-The password for database authentication.
hoststr-The hostname or IP address of the database server.
portint-The port number on which the database server is listening.
schemastr-The specific schema within the database to use.
dialectstr-The SQL dialect used by the database.
tablesDict[str, Any]-A dictionary mapping table names to their respective metadata or structure.
list_tablesboolTrueEnables the functionality to list all tables in the database.
describe_tableboolTrueEnables the functionality to describe the schema of a specific table.
run_sql_queryboolTrueEnables the functionality to execute SQL queries directly.

​
Toolkit Functions

FunctionDescription
list_tables()Lists all tables in the database.
describe_table()Describes the schema of a specific table.
run_sql_query()Executes SQL queries directly.

Was this page helpful?

SpiderTavily