test-code-review git:20260915.3de5d7aA

Immutable. This exact content is served forever at /api/v1/blob/ff10777ee03242e4.

---
name: test-code-review
description: Use this skill when you need to review test code for assertions, isolation, determinism, fixtures, and maintainability; triggers include 测试代码评审 and test code review.
---

# Test Code Review

identify assertion, isolation, fixture, determinism, and maintainability risks from test code, run evidence, and change scope. Produce TCR-## findings. This Skill organizes traceable test-engineering quality candidates only; it does not execute tests or turn a design inventory into coverage, pass, or release evidence.

## When to Use

- When you need test code review candidates from test code, fixtures, assertions, dependency injection, run history, failure logs, change diffs, and CI configuration.
- When you need selection rationale, applicability constraints, evidence gaps, and the smallest validation action.
- When inputs are incomplete but a bounded first pass can preserve blocked or unassessed boundaries.

Do not use it to execute tests, invent behavior conclusions, replace a complete strategy, or accept risk for a Human.

## Output Format Options

- Use Markdown by default; use tables, JSON, or CSV only when explicitly requested or required by the delivery format.
- Separate static analysis, unexecuted work, evidence states, and Human decisions; keep items unassessed, blocked, or NOT_RUN when runtime evidence is absent.

## How to Use

1. Read prompts/test-code-review.md and provide the objective, scope, material, environment, and evidence.
2. Complete the known, missing, conflicting, stale, out_of_scope, and assumptions input audit before findings.
3. Record TCR-## with the subject, preconditions, concern, source evidence, and validation, plus impact/priority, owner role, close condition, and evidence state.
4. Preserve conflicts, unknown constraints, and open questions when evidence is incomplete.

## Core Constraints

- Do not execute tests, assume missing rules, versions, thresholds, data, or outcomes, or treat candidate counts as coverage proof.
- File presence, names, design declarations, and Eval configuration are not runtime evidence.
- Mark unknowns unassessed, blocked, or pending clarification instead of filling them with convention.
- Do not edit requirements, code, test assets, or target systems.

## Pre-delivery Check

- [ ] Recorded the known, missing, conflicting, stale, out_of_scope, and assumptions input audit.
- [ ] Every TCR-## has source, evidence state, impact/priority, owner role, close condition, and validation.
- [ ] Facts, inferences, recommendations, unexecuted work, and Human decisions remain separate.
- [ ] Findings are not execution results, coverage proof, or release claims.

## Reference Files

- Read evals/eval.yaml and matching cases for regression; configuration does not prove project results.
- Use evals/trigger-prompts.csv and evals/local-rules.json for trigger checks; missing skill.selection evidence is BLOCKED.

## Common Pitfalls

- Do not turn a method name, file presence, or candidate count into test execution, coverage, pass, or release evidence when scope or evidence is incomplete.
- Do not fill in missing rules, thresholds, data, environments, or results from convention; preserve unassessed, blocked, and pending items.
- Do not expand this specialist design or review into a complete strategy, full test cases, runtime execution, or a release decision.

## Best Practices

- Complete the six-part input audit before selecting the smallest traceable and verifiable finding scope.
- Keep the source, evidence state, impact/priority, owner role, close condition, validation method, and residual risk for every finding.
- Write validation suggestions as next actions; do not upgrade package structure, candidate counts, or local Eval configuration into real quality conclusions.