From fly-ops
Deploy a Fly.io-hosted app safely and verify the rollout. Trigger on "задеплой", "deploy", "выкати на прод", or after merging work that must go live. Covers the push-to-main pipeline, the local flyctl fallback, the release_command gotcha, and post-deploy verification.
How this skill is triggered — by the user, by Claude, or both
Slash command
/fly-ops:deployThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
For apps wired to CI (racemodel: deploy = push to main), the deploy IS the
For apps wired to CI (racemodel: deploy = push to main), the deploy IS the push. Run the pre-push-review gate first, push, then watch CI through to the Fly release.
fly deploy -a <app> --remote-only
fly status -a <app> # all machines should reach `started`
Known gotcha: machines can get stuck in state created when the app has
a release_command (seen on racemodel). Fix: deploy via local flyctl without
the release_command (comment it out of fly.toml for the deploy, run the
migration manually), then restore.
fly status -a <app>: every machine started, latest version.fly logs -a <app> for the first minute: no crash loops, no tracebacks.fly releases -a <app> then fly deploy --image <previous-image>.Guides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.
Dispatches multiple subagents concurrently for independent tasks without shared state. Use when facing 2+ unrelated failures or subsystems that can be investigated in parallel.
npx claudepluginhub kulykivska/claude-plugins --plugin fly-ops