Load Azure AI Foundry data to DuckDB

Azure AI Foundry is a unified platform for enterprise AI operations, model builders, and application development that enables the creation and management of AI resources and generative AI applications. Everything needed to build a working Azure AI Foundry → DuckDB pipeline is on this page: the API's base URL, authentication, endpoints, pagination and incremental field — plus a prompt that hands the whole job to your coding agent.

---

Build your Azure AI Foundry to DuckDB pipeline

Paste this prompt into Claude, Codex, or Cursor. The agent does the rest.

PromptCopy prompt
Run uvx dlthub-init@latest to build a pipeline from Azure AI Foundry to DuckDB and run it on dltHub

That scaffolds a dltHub workspace and installs the dltHub AI harness — the project rules, the secrets-management skill, and the dlt MCP server your agent needs to work safely. From there it reads the Azure AI Foundry API, proposes the endpoints to load, then writes, runs and validates the pipeline while you review rather than type. Credentials are inspected through MCP tools, so your agent never reads secrets.toml itself. How the LLM-native workflow works →

Prefer to write it yourself? Every fact the agent uses is below.

---

Azure AI Foundry API at a glance

These values come from the Azure AI Foundry API reference — the authoritative source if anything here looks out of date.

---

How do I authenticate with the Azure AI Foundry API?

Azure AI Foundry supports Microsoft Entra ID (Bearer token) or API key authentication. For Entra ID, the Authorization header must be formatted as 'Bearer {token}', while for API keys, use the 'api-key' header.

1. Get your credentials

1. Navigate to the Azure AI Foundry portal and select your specific project. 2. Go to the Overview, Settings, or Keys and endpoints section. 3. Locate the Keys and endpoints area (or Connected resources). 4. Copy your API key (Key1 or Key2). Note: If not visible, ensure you have sufficient permissions (e.g., Owner, Contributor, or Cognitive Services Contributor) and that the key is stored in the associated Azure Key Vault.

2. Add them to .dlt/secrets.toml

Copy
[sources.azure_ai_foundry_source]
api_key = "your_api_key_here"

dlt reads this file automatically at runtime. With the harness, the setup-secrets skill prompts you for the values and never handles the raw credential in chat. For production, see setting up credentials with dlt.

---

What Azure AI Foundry data can I load into DuckDB?

These are the Azure AI Foundry endpoints dlt can load into DuckDB:

---

How do I load only new Azure AI Foundry records?

Azure AI Foundry exposes after on assistants?api-version=v1, so dlt can request only the records that changed since the last run. Set it as the cursor_path and dlt tracks the high-water mark for you between runs.

Copy
{"name": "agents", "endpoint": {
    "path": "assistants?api-version=v1",
    "data_selector": "data",
    "incremental": {"cursor_path": "after", "initial_value": "2024-01-01T00:00:00Z"},
}}

On the first run dlt loads everything from initial_value; on every run after that it requests only what changed and appends with write_disposition="merge" if you set a primary key. See incremental loading.

---

What does the generated Azure AI Foundry pipeline look like?

A standard dlt REST API pipeline — the same code you would write by hand, loading inference/chat and deployments from the Azure AI Foundry API into DuckDB:

Copy
import dlt
from dlt.sources.rest_api import RESTAPIConfig, rest_api_resources

@dlt.source
def azure_ai_foundry_source(api_key=dlt.secrets.value):
    config: RESTAPIConfig = {
        "client": {
            "base_url": "https://{resource-name}.services.ai.azure.com",
            "auth": {"type": "bearer", "token": api_key},
        },
        "resources": [
            {"name": "agents", "endpoint": {"path": "assistants?api-version=v1", "data_selector": "data"}},
            {"name": "evaluations", "endpoint": {"path": "evals?api-version=v1", "data_selector": "data"}}
        ],
    }
    yield from rest_api_resources(config)


def load_azure_ai_foundry_to_duckdb() -> None:
    pipeline = dlt.pipeline(
        pipeline_name="azure_ai_foundry_pipeline",
        destination="duckdb",
        dataset_name="azure_ai_foundry_data",
    )
    load_info = pipeline.run(azure_ai_foundry_source())
    print(load_info)


if __name__ == "__main__":
    load_azure_ai_foundry_to_duckdb()

Run it with python azure_ai_foundry_pipeline.py. The agent iterates on this until it loads cleanly — you review and approve, rather than write it from scratch.

---

How do I query Azure AI Foundry data in DuckDB?

dlt creates one table per resource. Query the loaded data with Python or SQL — or ask your agent to, through the MCP server's execute_sql_query tool.

Python (pandas DataFrame):

Copy
import dlt

data = dlt.pipeline("azure_ai_foundry_pipeline").dataset()
df = data.agents.df()
print(df.head())

SQL:

Copy
SELECT * FROM azure_ai_foundry_data.agents LIMIT 10;

See querying your data with dataset and exploring it in marimo notebooks.

---

How do I deploy the Azure AI Foundry to DuckDB pipeline in production?

The pipeline runs locally, which is ideal for prototyping and one-off analysis. When you need it on a schedule, monitored on every load, and shared with your team, deploy the same dlt code on the dltHub platform — no infrastructure to maintain. The prompt above already ends with "run it on dltHub", so your agent can take it there directly.

• Deploy & schedule — run the pipeline as a managed job with automatic retries.
• Monitor — observable job queues, alerting, and load metrics for every run.
• Transform — promote raw Azure AI Foundry loads into governed, documented models.
• Visualize & share — explore data in notebooks and publish live dashboards instead of static screenshots.

Book a demo →

---

What other destinations can I load Azure AI Foundry data to?

dlt loads into any of these — only the destination argument changes:

Set dlt.pipeline(destination="snowflake") and add credentials in .dlt/secrets.toml. On the dltHub platform the same pipeline runs against a managed Iceberg lakehouse. See the full destinations list.

---

Next steps

• Ready for production? Book a demo of the dltHub platform →
• dlt LLM-native workflow docs
• REST API source reference
• Join the dlt community on Slack