Load Promptfoo data to DuckDB

Promptfoo is a tool for evaluating and testing LLM prompts that provides a REST API for managing evaluations and user authentication. Everything needed to build a working Promptfoo → 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 Promptfoo 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 Promptfoo 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 Promptfoo 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.

---

Promptfoo API at a glance

These values come from the Promptfoo API reference — the authoritative source if anything here looks out of date.

---

How do I authenticate with the Promptfoo API?

Authentication is performed by including a static Bearer token in the 'Authorization' header of each request.

1. Get your credentials

1. Navigate to https://promptfoo.app/welcome to create an account or sign in. 2. Once authenticated, locate your account settings or profile section within the web interface. 3. Find the 'CLI Login Information' section to retrieve your API token. For Enterprise instances, your key can be accessed via the same section in your organization's specific instance URL.

2. Add them to .dlt/secrets.toml

Copy
[sources.promptfoo_source]
token = "your_bearer_token_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 Promptfoo data can I load into DuckDB?

These are the Promptfoo endpoints dlt can load into DuckDB:

---

How do I load only new Promptfoo records?

Promptfoo exposes offset on api/v1/me, 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": "me", "endpoint": {
    "path": "api/v1/me",
    "incremental": {"cursor_path": "offset", "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 Promptfoo pipeline look like?

A standard dlt REST API pipeline — the same code you would write by hand, loading /api/v1/me and /api/v1/users/me/teams from the Promptfoo API into DuckDB:

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

@dlt.source
def promptfoo_source(api_key=dlt.secrets.value):
    config: RESTAPIConfig = {
        "client": {
            "base_url": "https://api.promptfoo.app",
            "auth": {"type": "bearer", "token": api_key},
        },
        "resources": [
            {"name": "me", "endpoint": {"path": "api/v1/me"}},
            {"name": "teams", "endpoint": {"path": "api/v1/users/me/teams"}}
        ],
    }
    yield from rest_api_resources(config)


def load_promptfoo_to_duckdb() -> None:
    pipeline = dlt.pipeline(
        pipeline_name="promptfoo_pipeline",
        destination="duckdb",
        dataset_name="promptfoo_data",
    )
    load_info = pipeline.run(promptfoo_source())
    print(load_info)


if __name__ == "__main__":
    load_promptfoo_to_duckdb()

Run it with python promptfoo_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 Promptfoo 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("promptfoo_pipeline").dataset()
df = data.me.df()
print(df.head())

SQL:

Copy
SELECT * FROM promptfoo_data.me LIMIT 10;

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

---

How do I deploy the Promptfoo 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 Promptfoo 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 Promptfoo 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