Operations Orchestrator
The operational control plane for modern engineering teams
Coordinate releases, incidents, and operational workflows across your existing tools - with context, approvals, and traceability.
Runs locally. No production credentials required.
What It Does
Ops work doesn't live in one tool
Releases, incidents, investigations, and routine checks all require decisions across multiple systems. OpsOrch coordinates these workflows without replacing the tools you already use.
Signals (logs / metrics / alerts)
OpsOrch (context + decisions)
Actions (runbooks / approvals)
Back to tools (Grafana, Datadog, Jira, Argo)
One control plane. Many tools.
Designed For Real Work
Designed for real operational work
- Default views load with real data
- Copilot answers are inspectable and traceable
- Actions run through explicit workflows, not blind automation
Example
A typical flow
"Why did latency spike after the last change?"
Copilot correlates signals and shows evidence
A vetted runbook is suggested
The action runs with approvals and checkpoints
Open
Open by default
OpsOrch is built in the open. Core APIs, adapters, Console, MCP, and Copilot are open source. Browse the repos ↗
From The Blog
Operations Are Fragmented
An alert fires in Datadog. Investigation moves to Grafana. Decisions happen in Slack. The rollback runs through Argo. Follow-up lands in Jira. Every tool works. The process between them does not.