Hookdeck is an event gateway. It receives webhooks at a source URL, turns each request into one event per matching connection, and delivers events to a destination with queueing, rate control, and up to 50 automatic retries. A connection's rules can filter, transform, delay, or deduplicate events before delivery. Hookdeck's REST API at api.hookdeck.com is versioned by date, 2026-09-01 at present, and exposes requests, events, attempts, issues, and bulk retry jobs, together with the sources, destinations, and connections that route them.
Through Fibric, an operator reads failed events and open issues and proposes a retry, a pause on a connection, or a bulk replay. You approve, and the connector keeps the receipt with the bulk job id.
This is a reference listing. It documents what Fibric would read from Hookdeck and what it could propose, based on the vendor's published interfaces. Fibric builds it under a managed deployment when you request it; selecting it here installs nothing.
Inputs
Inbound requests from GET /requests, with GET /requests/{id}/events and /ignored_events showing which events each request produced or filtered out
Events from GET /events with a status of QUEUED, SCHEDULED, HOLD, SUCCESSFUL, or FAILED, and the raw body from GET /events/{id}/raw_body
Attempts from GET /attempts with status SUCCESSFUL or FAILED and trigger INITIAL or MANUAL
Issues from GET /issues of type delivery, transformation, backpressure, or request, and the total from GET /issues/count
Metrics from GET /metrics/events, /metrics/attempts, /metrics/queue-depth, and /metrics/events-pending-timeseries
Connections, sources, and destinations with disabled_at and paused_at timestamps and a destination's delivery_policy rate and period
Progress of bulk jobs from GET /bulk/events/retry/{id}, including completed_count and failed_count
Proposed actions
Target capability: propose retrying one event with POST /events/{id}/retry, or muting or cancelling a pending one with PUT /events/{id}/mute or /cancel
Target capability: propose a bulk retry of events matching a query with POST /bulk/events/retry, after reading estimated_count from GET /bulk/events/retry/plan
Target capability: propose pausing delivery on a connection with PUT /connections/{id}/pause, which marks new events HOLD, and resuming with /unpause
Target capability: propose replaying an inbound request through the pipeline with POST /requests/{id}/replay
Target capability: propose disabling or enabling a source or destination with PUT /sources/{id}/disable or PUT /destinations/{id}/enable
Proposed actions are target capabilities. Every action runs propose-first and needs a validated deployment and the appropriate permissions.
What you can build
Replay a day of failed deliveries after the fix ships
The operator reads events with status FAILED for the destination, gets estimated_count from the retry plan, and proposes a bulk retry. You approve. The receipt links the bulk job and its completed_count.
Hold a connection while the downstream system is down
A delivery issue opens for a destination. The operator proposes PUT /connections/{id}/pause so new events queue as HOLD, and the unpause once the destination answers again.
A Hookdeck project with sources, connections, and destinations already routing your webhooks
An API key from the project's Secrets page in the Hookdeck dashboard
The dated API version in every request path, for example /2026-09-01/events
For webhook notifications about issues, PUT /notifications/webhooks with enabled, topics, and the source_id that receives them
Authentication
A Hookdeck project API key from Settings, Project, Secrets in the dashboard, sent as Authorization: Bearer; Basic auth with the key as username is deprecated.
Limits
The API allows 240 requests per minute per API key and answers with Retry-After, X-RateLimit-Limit, X-RateLimit-Remaining, and X-RateLimit-Reset headers
List endpoints return 100 items by default and at most 250 per page, ordered by created_at descending unless order_by and dir say otherwise
Automatic retries stop at 50 per event, and a retry can be scheduled at most 7 days ahead. Manual retries are unlimited
Inbound payloads are capped at 10 MiB by default and searchable payload size at 2.5 MB. Developer plan throughput defaults to 5 events per second
Access and pricing
Reference listing. Fibric builds the connector under a managed deployment when you request it. Your quote covers the build, capabilities, usage, and support.
A connection's retry rule sets strategy (linear or exponential), interval in milliseconds, count, and response_status_codes, with ranges such as 400-499 and exclusions such as !401. By default any non-2xx response triggers a retry. Automatic retries are limited to 50, and a Retry-After header can schedule one up to 7 days ahead.
What is the difference between a request, an event, and an attempt?
A request is the inbound webhook Hookdeck received at a source. Hookdeck creates one event per connection that request matches; filtered-out ones appear under ignored_events. An attempt is one delivery of an event to its destination, recorded with status SUCCESSFUL or FAILED and trigger INITIAL or MANUAL.
How is the Hookdeck API versioned?
By date in the path: https://api.hookdeck.com/2026-09-01/ is the latest. Earlier versions such as 2025-07-01 and 2025-01-01 stay available for about a year after release. List endpoints paginate with next and prev cursors, a limit of up to 250, order_by, and dir.
Ask about Hookdeck
Ask about the capabilities and requirements in this listing.
This connector listing is developed, published, and supported by Fibric. It describes integration with Hookdeck through published interfaces. Third-party names and logos identify the systems an integration connects to; they are the property of their respective owners, who are not affiliated with Fibric and do not sponsor or endorse this listing. Trademark policy