Events and outcome reporting
Connect signed outbound events and inbound triggers to systems outside CodeBook.
CodeBook records events for workflow activity and can deliver signed webhooks to connected systems. Inbound triggers can open work or report supported outcome signals, such as a production issue linked to finished work.
Outbound delivery
Configure the destination and its event subscriptions. The receiving service must validate the signature and handle repeated deliveries safely. A delivery failure should be investigated using its recorded status; it does not prove that the underlying run failed.
Send only the events your integration needs. Keep signing credentials in the connection configuration rather than in card descriptions or agent skills.
Inbound triggers
Connect a monitoring or reporting system to the supported authenticated trigger path. Map its event to the correct factory, station or outcome. Validate the mapping with a test event before relying on it for incident handling.
Outcome signals are useful only when they are connected and correctly attributed. No incoming report is not the same as a confirmed healthy deployment.
The inbound trigger address, like every address CodeBook hands to another system — monitoring sources, external runtimes' brief and callback URLs, and the git webhook — answers at CodeBook's web address.
Dedicated external queue-broker adapters are unfinished. Use the existing webhook or MCP contracts where they fit, and consult implementation status before planning around an additional adapter.
Production monitoring
Incidents and deploys from Sentry, Datadog, PagerDuty, Opsgenie, Grafana / Alertmanager or a generic webhook arrive through monitoring sources on the System Map rather than through triggers: incidents become incident nodes linked to runtimes, services and modules, and deploys record which commit a runtime is running. A change of an incident's status publishes incident.opened or incident.resolved in the System map group, beside impact.computed, to every factory with a workflow attached to the linked node's repository. See Production monitoring.