Blog
Notes on reliable event infrastructure.
Product updates, architecture notes, and workflow patterns for teams building on Hooksbase.
Webhook DLQs: design and recovery patterns
A DLQ holds messages that failed terminally so you can inspect and recover them. Here is how DLQs work, how to size the retry budget, what to put in each entry, and the five recovery patterns that ship in production.
Webhook testing: how to test webhooks locally, in CI, and in production
Testing webhooks is awkward because they come from outside your machine, on someone else's schedule, with payload shapes you don't always control. Here's the right tool for each stage — local, CI, and production debugging.
The complete guide to idempotency keys (for webhooks and APIs)
An idempotency key is a unique value attached to a request that tells the receiver: if you've seen this key before, return the same response without doing the work again. Here is how it works in practice — and what breaks without it.
Routing, transforms, and replay for AI agents
Event routing and payload transforms are only useful when failed agent events can be replayed deterministically — with the exact payload and destination they would have reached the first time.
Deterministic replay is non-negotiable for agents
A classical webhook consumer is deterministic. An agent isn't. That changes what 'replay' has to mean — and it's the one feature most agent stacks quietly get wrong.