Migrate from Hookdeck to Hooksbase

This guide is for teams on Hookdeck considering Hooksbase — usually because the metered event pricing stops making sense at scale, or because the workflow has shifted to serving an AI agent that needs richer replay and typed outbound destinations.

See the Hooksbase vs Hookdeck comparison if you're still evaluating.

Concept mapping

Pricing shift

If your current Hookdeck plan is metered or event-volume-sensitive, compare it against Hooksbase's bundled delivery tiers ($0, $25, $79, $249, custom).

Bundled tiers are more predictable than pure usage pricing, but the cost crossover depends on your traffic profile and current contract. Pull a recent month of event volume from Hookdeck, apply the current pricing from your plan, and compare to the closest Hooksbase tier.

If pure per-event economics matter more than predictability, your existing Hookdeck plan may still be the right fit.

Transformation mapping

Hookdeck transformations use JavaScript. Hooksbase payload transforms use JSONPath (for extraction) or Handlebars JSON templates (for templated shape).

For simple field extraction, the translation is direct:

// Hookdeck
$.request.body.customer.email

// Hooksbase JSONPath
$.customer.email

For complex transformations that use arbitrary JavaScript, you'll need to:

1. Extract the transformation logic into your agent (do it in the agent code, not the event layer)
2. Or use Handlebars JSON templates for simple templating
3. Or keep a minimal Worker proxy that does the JS transformation and calls Hooksbase

Most agent-oriented flows move transformation logic into the agent itself, which keeps the event layer simpler and the agent in control of its input shape.

Connection translation

A Hookdeck connection is a source + destination + filter. In Hooksbase:

1. Create the webhook (the ingest endpoint + default destination + signing config)
2. Add additional destinations for routing different event classes if needed; for true fan-out of the same event, route to a downstream broker or fan-out service
3. Define routing rules matching Hookdeck's filters

Hookdeck filter syntax (JSONPath/JavaScript) maps to Hooksbase routing conditions:

Switchover plan

1. Create the webhook in Hooksbase with the same destination URL as Hookdeck's connection
2. Configure routing rules to match Hookdeck's filter logic
3. Migrate transformations (JSONPath/Handlebars or into the agent)
4. Dual-run: send to both Hooksbase and Hookdeck from low-volume producers first
5. Verify parity: same deliveries, same agent behavior, same observability
6. Cut over: update remaining producers to Hooksbase
7. Decommission Hookdeck after a 14-day retention window

What to carry over

• Signing secrets (if the consumer was doing its own verification against your signed outbound)
• Destination URLs — unchanged
• Retry semantics — rebuild in Hooksbase custom retry policy (Starter+)

What you gain

• Bundled pricing that scales predictably
• Multi-channel ingest (email and forms alongside HTTP; scheduled cron on Starter+)
• Built-in file relay for email and form attachments on paid tiers
• Starter+ provider packs (Stripe, GitHub, Clerk, Slack, Resend) with pre-persistence verification
• Deterministic dispatch snapshots so replays survive transform config changes
• Pro+ event drains for observability

What Hookdeck still does better

Be honest about the trade:

• More granular event-count metering at scale
• More mature alerting/issue composition surface
• Broader metrics export options

If those are critical for your workload, Hookdeck may still be the right call. Hooksbase makes sense when the consumer is an AI agent and when bundled packaging aligns better with your spend profile.

What next

• Compare Hooksbase vs Hookdeck
• Verify provider webhooks before they reach your agent