How do I make a screening agent resilient with a fallback source?

Steps

1. Install the Strale Python SDK, straleio.
2. Wrap your primary screening call in a try/except.
3. On failure, call sanctions-check with the same name.
4. Tag the result as a fallback so downstream systems know which source answered.

The solution

# pip install straleio
from straleio import Strale, DoRequest

strale = Strale(api_key="sk_live_YOUR_KEY")


def screen_entity(name: str) -> dict:
    """Screen with the primary source, and fall back to Strale."""
    try:
        return primary_sanctions_api.check(name)  # your primary provider
    except Exception as err:
        result = strale.do(DoRequest(
            capability_slug="sanctions-check",
            inputs={"name": name},
            max_price_cents=30,
        ))
        return {
            "source": "strale_fallback",
            "matches": result.output.get("matches", []),
            "fallback_reason": str(err),
        }


print(screen_entity("Acme Corporation"))

How it works

Screening that runs on every onboarding should not stop because one provider is down. The pattern tries your primary provider first and switches to Strale on a timeout, rate limit or error.

Tag fallback results with their source, as source does here, so your records show which provider produced each answer.

Before relying on it, run both sources on names where you know the expected answer, so you understand how they differ before an outage forces the switch. Either way, matches are for a person to review.

Tools in this guide