Does it support deep transitive dependencies?

It supports one level of transitive analysis to keep the 'Impacted' tier efficient without over-expanding the test surface.

How does it handle flaky tests?

It maintains a reliability log that tracks pass/fail history, automatically excluding unstable tests from the high-speed Tier 1 feedback loop.

What testing frameworks are supported?

This skill is currently optimized for Python projects using pytest and git for version control.

How does Smart Test determine which tests to run?

It analyzes your git diff to identify changed files and maps them to tests using a bidirectional import dependency graph.

When should I avoid using smart test selection?

Avoid using it during major refactors, database schema changes, or final pre-merge checks, as these scenarios require a full suite validation.

Smart Test Selection

Name: Smart Test Selection
Author: rysweet

byrysweet

•

セキュリティとテスト

Optimizes testing workflows by intelligently selecting affected tests based on code changes and reliability data.

概要

Smart Test Selection is a performance-focused skill for Claude Code that reduces feedback loops and accelerates local development. By performing import dependency analysis and tracking test reliability, it categorizes tests into three logical tiers: Fast, Impacted, and Full Suite. It automatically maps Python source files to their corresponding tests, identifies flaky tests to prevent false negatives in quick feedback loops, and provides ready-to-use pytest commands tailored to your current git diff.

主な機能

Automated import dependency mapping
Smart feedback loop optimization
Three-tier testing strategy (Fast, Impacted, Full)
17 GitHub stars
Flaky test tracking and reliability scoring
Seamless git integration for change detection

ユースケース

Validating PR drafts with impacted integration tests before a full suite run.
Accelerating local pre-commit checks by running only affected unit tests.
Identifying and quarantining flaky tests based on historical pass/fail data.

概要

主な機能

Automated import dependency mapping
Smart feedback loop optimization
Three-tier testing strategy (Fast, Impacted, Full)
17 GitHub stars
Flaky test tracking and reliability scoring
Seamless git integration for change detection

ユースケース

Validating PR drafts with impacted integration tests before a full suite run.
Accelerating local pre-commit checks by running only affected unit tests.
Identifying and quarantining flaky tests based on historical pass/fail data.