feature-flag-auditor

Audit codebases for stale, orphaned, and misconfigured feature flags.

3|2|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/robotijn/ctoc --skill feature-flag-auditor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: feature-flag-auditor
Source: https://github.com/robotijn/ctoc/tree/main/skills/versioning/feature-flag-auditor
Command: npx skills add https://github.com/robotijn/ctoc --skill feature-flag-auditor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Feature flags that are left unowned, lack sunset dates, or remain at 100% rollout long after deployment accumulate as hidden technical debt that causes production incidents, failed rollbacks, and wasted engineering effort. This skill automates flag audits to catch these hygiene issues before they escalate to customer-facing outages.

Core Features & Use Cases

  • Stale Flag Detection: Identifies flags fully rolled out for 30+ days, unused for 90+ days, or missing required ownership and sunset metadata.
  • Risky Path Protection: Flags missing kill-switches for high-risk code paths including payments, PII handling, LLM calls, and external API integrations.
  • Cross-Platform Support: Works with OpenFeature, LaunchDarkly, Statsig, PostHog, Flagsmith, Unleash, GrowthBook, ConfigCat, and custom flag implementations.
  • Use Case: An engineering team can run this audit pre-release to catch orphaned flags and missing kill-switches that would otherwise cause production fires during rollbacks or incident response.

Quick Start

Use the feature-flag-auditor skill to scan your codebase for stale, unowned, or improperly configured feature flags and generate a prioritized cleanup report with remediation steps.

Frequently Asked Questions about feature-flag-auditor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect stale feature flags in my codebase?

Detecting stale feature flags requires scanning your codebase to identify toggles fully rolled out for 30+ days, unused for 90+ days, or missing required ownership and sunset metadata. This prevents hidden technical debt from causing production incidents.

Why does my feature flag need a kill switch for risky code paths?

Feature flags need kill switches for risky code paths like payments, PII handling, LLM calls, and external API integrations to enable rapid rollback. Missing kill-switch coverage causes production fires and failed rollbacks during incident response.

Does the feature flag audit work with LaunchDarkly and OpenFeature?

Yes, the feature flag audit works with OpenFeature, LaunchDarkly, Statsig, PostHog, Flagsmith, Unleash, GrowthBook, ConfigCat, and custom implementations to detect unowned or improperly configured toggles across your codebase.

What is feature flag hygiene and why is it important?

Feature flag hygiene is the practice of enforcing flag ownership, sunset date policies, and kill-switch coverage to eliminate stale toggles. It is important because unmanaged flags accumulate as hidden technical debt, causing failed rollbacks and wasted engineering effort.

How do I enforce both-branch test coverage for flag-controlled logic?

To enforce both-branch test coverage for flag-controlled logic, run a pre-release validation audit that checks for missing kill-switches and improperly configured toggles. This generates a prioritized cleanup report with specific remediation steps.

Can I audit custom feature flag implementations or only vendor platforms?

You can audit both vendor platforms and custom feature flag implementations. The audit detects orphaned flags and missing sunset metadata across OpenFeature, commercial platforms, and your own custom flag logic.