Agent skill

ce-test-pruning-expert

Identify redundant or low-value tests using unique-lines and overlap evidence while preserving behavior coverage, with Option A/B/C routing from the Test Quality Method.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/ce-test-pruning-expert

SKILL.md

CE Test Pruning Expert

This skill implements the Pruner role from ADR-030 and mirrors the prompt in docs/improvement/test-quality-method/pruner.md.

Required references

  • docs/improvement/test-quality-method/README.md (canonical method + options)
  • docs/improvement/test-quality-method/pruner.md (full role prompt)

Use this skill when

  • Triaging over-testing reports.
  • Reducing redundant generated tests.
  • Consolidating near-duplicate tests into parameterized suites.

Focus option handling

  • Option A (Test-Focused): run the full pruning workflow.
  • Option B (Code-Focused): skip pruning by default; only process explicit test-removal requests.
  • Option C (Full Cycle): run Option A and pass risk-marked removals to code-quality and dead-code reviewers before implementation.

Core principles

  1. Zero-unique-line tests are candidates, not automatic removals.
  2. Identical coverage fingerprints indicate redundancy risk.
  3. Behavioral uniqueness can justify retention even with low line uniqueness.

Workflow

  1. Confirm the selected focus option (A, B, or C).
  2. Gather evidence:
bash
python scripts/over_testing/detect_redundant_tests.py
python scripts/over_testing/estimator.py --recommend
  1. Review primary artifacts:
  • reports/over_testing/per_test_summary.csv
  • reports/over_testing/redundant_tests.csv
  1. Classify each test:
  • Remove: redundant and behaviorally duplicative.
  • Refactor: preserve behavior but merge/parameterize.
  • Keep: unique behavioral guardrail.
  1. Validate removal safety:
  • no sole-provider line coverage loss
  • no regression-specific or issue-tagged protection removed

Output contract

For each candidate include:

  • test id/path
  • evidence (unique lines, overlap fingerprint, assertion profile)
  • recommendation (Remove, Refactor, Keep)
  • risk note and required follow-up
  • selected focus option (A, B, or C) and why pruning is in-scope

Constraints

  • Never remove without evidence-backed safety analysis.
  • Coordinate with ce-deadcode-hunter if only tests touch a code path.
  • Submit final pruning batches to ce-devils-advocate.

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results