I built OpsOrch because operational work is spread across too many tools.
Incidents might be in PagerDuty, logs in Datadog, metrics in Prometheus, and runbooks somewhere else. Each system has its own API and data model, so cross-tool workflows usually turn into glue code.
OpsOrch is my attempt to standardize that.
It defines a common schema for things like incidents, logs, metrics, alerts, services, and runbooks, then uses adapters to connect different providers behind one interface.
The goal is simple: instead of writing vendor-specific logic, a client can ask for:
- incidents affecting a service
- related logs and metrics
- associated runbook context
I am especially interested in feedback on whether this abstraction is actually useful, or whether it becomes too generic in practice.
Happy to answer questions about the schema, adapter model, and tradeoffs.
No comments yet.
