Ask whether a webhook can arrive twice

Format: Explainer.

A webhook is a message sent by another system when something happens. Getting that message does not always mean there is new work to do. The tool that links your systems must be able to spot the same delivery twice.

Shopify documents that an app may receive the same webhook more than once, including after a timeout or retry.[10] That is a provider example, not a claim that your current tools are connected.

Try this hypothetical replay test with made-up records. A message says internal request R-17 is ready for review. The proposed tool creates one review task. Send the same approved test message again. There should still be one task. Ask to see where the tool records that it caught the repeat.

Keep a short test record: source account; event ID; first result; repeat result; saved task ID; anything still unknown. Include the account with the event ID. The tool must not mix up two customers that happen to use the same ID.

Then ask what happens if the task was created but the connector stopped before saving its success status. A “seen it” list may not answer that. Ask how the tool keeps the task's ID and checks whether it already exists before trying again. A second task is not a safe way to clear an unknown result.

Ask: “Can you show one saved task after both the repeat and the restart?” A success reply to the sender is not enough. Passing this test does not prove that all work will happen exactly once. Check each provider and action the build will use.

Book a Demo with Core Agentic to discuss duplicate handling for a relevant custom integration, using approved test inputs rather than live customer actions.

Sources

• Shopify: Verify webhook deliveries.

These sources describe the named provider or framework. They do not establish a Core Agentic integration, customer result or certification.