-
Notifications
You must be signed in to change notification settings - Fork 1
Operations and Runbooks
Mac Anderson edited this page Jul 16, 2026
·
1 revision
Deploying, monitoring, and fixing Stella in production. If you're on call, this page and its children are your map.
🚧 Starter page — every incident should leave this page better than it found it.
Environments, deploy commands/pipeline, rollback procedure, and who can deploy.
| Signal | Dashboard | Alert threshold | First response |
|---|---|---|---|
<latency, error rate, token spend, sandbox health…> |
<link> |
<value> |
<runbook link> |
Create one page per scenario and link it here (naming: Runbook-<Scenario>):
-
Runbook-LLM-Provider-Outage(todo) -
Runbook-Sandbox-Escalation(todo) -
Runbook-Rollback(todo)
Severity levels, escalation path, comms channel, and postmortem process. Postmortems live under Decision Records or a dedicated Postmortem-YYYY-MM-DD-<slug> page.
- Architecture — what you're operating
- Development Guide — how fixes get shipped
Stella Engineering Wiki · internal use only · How to contribute · Questions → #stella-eng · Sponsor me! https://github.com/macanderson - thanks!
Getting Started
Build
Operate
Reference
curl -fsSL https://raw.githubusercontent.com/oxageninc/stella/main/install.sh | sh
stella --version- Oxagen: https://oxagen.sh
- Stella Docs: https://stella.oxagen.sh
- X: https://x.com/macandersoncto
- LinkedIn: https://linkedin.com/in/macanderson
- Email: mac@oxagen.sh