Featurevisor
Feature flags and remote config management with GitOps
FreeWorkflow automationDevOpsOpen Source15,969 impressions#89 of its week
Featurevisor provides Git-based feature flags, A/B tests, and remote configuration managed as code.
- for
- Developers and teams who want version-controlled feature management.
- pricing
- open source
- license
- MIT
Key features
- Git-backed definitions — Features, segments, and variables are stored as YAML files in your repository and changed via pull requests.
- Static datafile deployment — CI builds and uploads JSON datafiles to a CDN, eliminating runtime network calls per lookup.
- Gradual rollouts & targeting — Deterministic bucketing and segment rules enable percentage rollouts and audience targeting.
- Typed variables & overrides — Global variables support schemas, defaults, and hierarchical overrides validated at build time.
- A/B experiments — Define variations with weighted traffic splits and retrieve the assigned variation at runtime.
- Environment & promotion sets — Manage multiple environments or sets in the same repo and promote changes through CI.
Use cases
- Enable a new checkout redesign for Dutch users before a full rollout
- Run A/B tests on pricing pages with weighted traffic distribution
- Configure per-account limits via hierarchical variable overrides
Featurevisor FAQ
How are feature changes reviewed?+
Changes are made as pull requests, reviewed like code, and merged through the standard Git workflow.
Do lookups require network calls?+
No, the SDK loads the static JSON datafile once and evaluates flags in memory without per-lookup network requests.
Can I target specific user segments?+
Yes, you define attributes and segments in YAML and reference them in rules to target audiences.
How are environments handled?+
Rules can be scoped under environment keys (e.g., staging, production) within the same file, allowing separate rollout schedules.
Is Featurevisor compatible with AI agents?+
Featurevisor includes an official skill that lets AI agents author YAML definitions via natural-language commands.
Summarized by DevHunt from featurevisor.com · Oct 4, 2026. Details may change; check the official site.
.png?width=64)
