/Docs
IntroductionHow FeatureGate worksQuickstart
OrganisationsProjectsEnvironments
OverviewTypes and valuesConfigure across environmentsTargeting rulesPercentage rolloutsDesign evaluation contextLifecycle and intentTags and discovery
Choose an evaluation approachEvaluation credentialsNode.js SDKEvaluate over HTTPEvaluate in browsersSnapshots and configuration freshnessErrors and fallbacksTest flags locallyProduction integration checklist
Run a controlled rolloutKill switchesProject insightsActivity and audit historyReview and clean up flagsEnvironment readinessTroubleshooting and limits
Members, roles, and permissionsInvite and manage a teamOrganisation settings and operational defaultsProject lifecycle and transferPlans, billing, and team organisationsOrganisation closure and reactivationAccount settings and security
Management API overviewPreviewProvision flags through automationPreviewRotate runtime keys through automationPreviewAudit project changes through automationPreview
DocsAPI Reference

Flags/Tags and flag discovery

Get access
/Docs
Get access
  1. Docs
  2. Flags
  3. Tags and flag discovery

Tags and flag discovery

Build an organisation-wide vocabulary for finding and reviewing related flags.

MarkdownFeedback
PreviousFlag lifecycle and intentNextChoose an evaluation approach

On this page

Build a useful vocabularyCreate and attach tagsSearch and filterRuntime boundaryMaintain the vocabularyCommon problems

Tags are organisation-owned labels attached to flags. They help people find related controls across projects without changing runtime evaluation.

Use tags for recurring discovery questions such as ownership area, product surface, initiative, or service. Do not duplicate lifecycle and runtime state that FeatureGate already models directly.

Build a useful vocabulary

A small governed vocabulary is more useful than many near-duplicates.

Good tagQuestion it answers
checkoutWhich flags affect checkout?
identityWhich flags belong to identity work?
mobileWhich flags affect the mobile surface?
search-serviceWhich service owns these controls?

Avoid tags such as active, urgent, production, or cleanup-due. Status, environment, intent, and cleanup signals already answer those questions and remain accurate automatically.

Avoid employee names and raw customer names. Use stable team or account references only when the organisation has approved that taxonomy and understands its privacy implications.

Create and attach tags

Tags can be created inline during flag creation and editing. A flag can carry several tags. Tags with no remaining flags disappear from the active vocabulary.

Use the same spelling and granularity across projects. Before creating a new label, search for an existing synonym such as payments versus billing.

Members, admins, and owners can manage flag tags as part of normal project building. Viewers can use tags for discovery but cannot mutate them.

Search and filter

Free-text flag search matches tag labels as well as flag identity. Project flag lists can filter by one organisation tag key at a time.

Combine tags with dedicated filters. For example:

  • checkout plus Cleanup due finds temporary debt in one product area.
  • identity plus Partial rollout finds releases needing an owner decision.
  • search-service plus archived discovery helps verify code cleanup.

Do not encode several dimensions into one label such as checkout-production-cleanup. Use one stable area tag and let environment and lifecycle filters provide the changing state.

Runtime boundary

Tags are console metadata. They are not included in evaluation, snapshots, invalidation, or the public Management API preview.

Adding, removing, or renaming a tag cannot change a flag value or refresh an SDK. Automation that needs tags is not currently supported by the public Management API.

Maintain the vocabulary

Review tags during cleanup work or organisation ownership changes:

  1. Identify near-duplicates and spelling variants.
  2. Choose the surviving organisation term.
  3. Retag affected flags.
  4. Confirm search and saved operational habits still work.
  5. Allow unused tags to disappear.

Document a small set of preferred area tags for larger teams. Avoid building a deep taxonomy that requires constant administration.

Common problems

  • A tag is missing: it may have no remaining flags.
  • Search returns too much: use a tag filter with lifecycle or status filters.
  • A runtime value did not change: tags never participate in evaluation.
  • Automation cannot attach a tag: the current Management API preview does not expose that workflow.

Related pages: Flag lifecycle and intent, Review and clean up flags, and Activity and audit history.