Skip to content
dinobase

Pydantic AI

Dinobase provides a Pydantic AI toolset that lets your agents query business data from 100+ sources with full type safety and dependency injection.

Install

Section titled “Install”
Terminal window
pip install "dinobase[pydantic-ai]"

Set up your data sources:

Terminal window
dinobase init
dinobase add stripe --api-key sk_test_...
dinobase add hubspot --api-key pat-...
dinobase sync

See Connecting Sources for the full list of 100+ supported sources, and Syncing & Scheduling for background sync options.

Quick Start

Section titled “Quick Start”

Use the pre-configured agent:

from dinobase.integrations.pydantic_ai.tools import DinobaseDeps, dinobase_agent


result = dinobase_agent.run_sync(
    "Which customers have overdue invoices?",
    deps=DinobaseDeps(),
)
print(result.output)

Toolset on Your Own Agent

Section titled “Toolset on Your Own Agent”

Attach the dinobase_tools toolset to any Pydantic AI agent:

from pydantic_ai import Agent
from dinobase.integrations.pydantic_ai.tools import DinobaseDeps, dinobase_tools


agent = Agent(
    "anthropic:claude-sonnet-4-6",
    deps_type=DinobaseDeps,
    toolsets=[dinobase_tools],
    instructions="You are a financial analyst. Be concise.",
)


result = agent.run_sync("What is our MRR trend?", deps=DinobaseDeps())
print(result.output)

Tools

Section titled “Tools”

The dinobase_tools toolset provides four tools:

ToolDescription
dinobase_queryExecute SQL queries (DuckDB dialect)
dinobase_describeGet table schema, types, and sample data
dinobase_list_sourcesList connected sources with freshness status
dinobase_refreshRe-sync a stale data source

All tools use RunContext[DinobaseDeps] for type-safe dependency injection. The DinobaseDeps dataclass lazily initializes the Dinobase QueryEngine.

How It Works

Section titled “How It Works”

Tools wrap Dinobase’s Python API via Pydantic AI’s FunctionToolset. Dependencies are injected through RunContext[DinobaseDeps], which provides access to the QueryEngine.

The typical agent workflow:

  1. dinobase_list_sources — discover available data
  2. dinobase_describe — understand table schemas
  3. dinobase_query — run cross-source SQL queries
  4. Present and analyze results

Example

Section titled “Example”

See the example agent:

Terminal window
export ANTHROPIC_API_KEY=sk-ant-...
python examples/analyst.py "Which deals closed this quarter?"

Next steps

Section titled “Next steps”