Selects appropriate test strategies and layers for code changes based on risk.
npx skills add https://github.com/shiwenwen/hope-agent --skill ha-test-strategy
Please help me install this Agent Skill. Skill: ha-test-strategy Source: https://github.com/shiwenwen/hope-agent/tree/main/skills/ha-test-strategy Command: npx skills add https://github.com/shiwenwen/hope-agent --skill ha-test-strategy
Deciding which tests to write for a feature, bug fix, or refactor is often inconsistent and subjective. This Skill provides a structured decision framework that matches the right testing approach to the risk and nature of each code change.
Ask the agent to choose and add the right regression test for the bug you are fixing.
High-intent search queries and answers about installing and using this skill.
Use a regression-first approach: reproduce the original failure in the narrowest credible automated test, confirm it fails for the right reason, then verify it passes after the root-cause fix. This proves the fix addresses the actual bug.
Test-first works best when the new contract is clear, the test seam is stable, and observing the failure proves the test is meaningful. It is a valuable technique, not an unconditional law for every change.
Use the lowest layer that proves the contract: unit tests for pure logic, integration for persistence or cross-module contracts, and E2E only for user-critical paths spanning the full stack. Add higher layers only for boundaries lower layers cannot cover.
Use characterization-first testing: capture current intentional behavior before changing structure. Do not freeze a known bug as desired behavior—only preserve behavior that is genuinely correct.
Skipping new tests is reasonable for pure docs, trivial metadata, generated outputs, or low-risk mechanical changes when existing checks already cover the risk. Explain the decision rather than adding a meaningless test.
Investigate the root cause of the flakiness or environment issue instead of rerunning until green. Repeatedly rerunning a flaky test until it passes does not constitute proof of correctness.
pecorapromotionsllc-cellDrive development with failing tests before implementing logic.
mcsdodoRun tests, check git status, and confirm changelog updates for completion readiness.
robwestzAnalyze pull requests for correctness, security, performance, and style issues.
robotijnGenerate property-based tests validating universal invariants across full input spaces.
