Skip to content

scout

UI-driven API regression testing. Maintained by BoxProbe.

scout drives your web app's UI like a real user, records the resulting API traffic, and produces a diff report between two runs. When a deploy silently changes a response shape, scout tells you which endpoint changed, in which user flow, and exactly how.

Scenarios are recorded once and replay deterministically — no AI in the hot path, no per-request API fees, no SaaS subscription. Same scenario on the same app produces the same trace, at $0 per run.

Three pillars

Property How
Deterministic Pixel-anchored Locators make element resolution pure math, not a probabilistic selector match. No LLM runs in the hot path. Same input, same trace.
Near-zero cost No tokens consumed, no per-request API fees. Cheap enough to run on every PR or nightly without anyone asking about cost.
Auditable Open source, no telemetry, no upload step — everything stays on the machine running scout. Read the Python before letting it into your CI.

Where to start

  • Quickstart — install scout, write a scenario, see your first diff report
  • Scenarios — the Scenario, Locator, and Page APIs
  • Diff ignore rules — how to suppress timestamp / ID noise without losing real regressions
  • CLI reference — every subcommand and its options

Try it on a real app

github.com/boxprobe/scout-medusa is a fully reproducible demo: two pinned versions of Medusa running side by side, 15 admin scenarios, one HTML diff report. Clone, docker compose up, run scout twice, see the diff.

Status

scout is alpha (0.1.x). The CLI surface and scenario DSL are still stabilizing; expect breaking changes between minor versions until 1.0. See the changelog for what's shipped.