# Canary

> Post-deploy monitoring workflow for Codex. Use when a live deployment should be watched for obvious breakage, regressions, or unhealthy signals.

- Skill: `mphaxise/canary` (Agent Skill)
- Install (CLI): `npx skillmds@latest add mphaxise/canary`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mphaxise/canary/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- Author: mphaxise (https://skillmd.com/u/mphaxise)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/mphaxise/canary

---


# Canary

Use this skill after deployment when the right move is to watch the live app for regressions instead of assuming success.

This port is adapted from `garrytan/gstack` at commit `4d2c8d94d00cc4f4f3d4c26316a4f939ceedc045`.

## Workflow

1. Define the live surface to monitor.
2. Check the most important production signals:
   - page availability
   - console or runtime errors
   - key route health
   - performance drift
3. Record the result in a report if the run should be compared later.
4. If recurring checks are useful, hand off to `cron-scheduler`.

## Guardrails

- Do not treat one successful page load as comprehensive health.
- Be explicit about what was not monitored.
- Prefer short, repeatable checks over broad but vague post-deploy confidence.