cyberneticlibrary

Architect and manage feature flags

feature-flags-architectpluginsetup L217,464
alirezarezvani/claude-skills
What it does

Classify, ship, ramp, and retire feature flags with phased rollout and debt detection

Best for

When shipping features requires disciplined flag management with rollout rings, stale-flag cleanup, and audit trails rather than ad-hoc toggles.

Inputs
  • · feature definition: scope, rollout strategy (ring, linear, log, cohort)
  • · flag provider + configuration
Outputs
  • · flag classification: experiment, release, ops, etc.
  • · phased rollout plan with abort criteria
  • · stale flag detector (tracks debt)
  • · kill-switch audit (every flag has documented kill path)
Requires
  • · 3 stdlib Python tools (classifier, rollout planner, stale detector)
  • · 4 references: flag taxonomy + provider trade-offs (LaunchDarkly, GrowthBook, Statsig, Unleash, Flipt, DIY) + rollout strategies + lifecycle
  • · /flag-cleanup slash command
  • · provider integrations (agnostic)
Preconditions
  • · flag provider installed + configured
  • · rollout strategy chosen
  • · abort criteria defined
Failure modes
  • · Phased rollout too slow, misses market window
  • · Kill-switch audit finds flag with no documented retire path (becomes permanent debt)
  • · Flag taxonomy doesn't match team's actual experiments vs releases
Trust signals
  • · 3 stdlib Python tools (classifier, rollout planner, stale-flag detector)
  • · 4 deep references (taxonomy + provider comparison + rollout strategies + lifecycle)
  • · /flag-cleanup slash command
  • · Cross-tool compatible (works with LaunchDarkly, GrowthBook, Statsig, Unleash, Flipt, DIY)