Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add fishzjp/qa-skills --skill test-case-reviewgit clone --depth 1 https://github.com/fishzjp/qa-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/fishzjp/qa-skills/test-case-review)<a href="https://agentmods.dev/skills/fishzjp/qa-skills/test-case-review"><img src="https://agentmods.dev/badge/skills/fishzjp/qa-skills/test-case-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/fishzjp/qa-skills/test-case-review"><img src="https://agentmods.dev/badge/skills/fishzjp/qa-skills/test-case-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00109 | $0.01767 |
| Opus 5.5 | $0.00044 | $0.00707 |
| Sonnet 5.5 | $0.00022 | $0.00353 |
| Haiku 4.5 | $0.00011 | $0.00177 |
Grade A, and why
test-case-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
测试用例审查(test-case-review)
回答"这些测试用例到底测得好不好"——事后、独立的审查(写时自审归 test-case-writing 阶段四)。
- 输入:已有用例文件(markmap + Schema,若无可先行抽取)、PRD / 需求模型、代码仓库
- 输出(落盘):直接修订用例文件(修订后重新抽取 Schema)+ 审查记录(文件末尾附录)
- 审查记录内容:缺失 / 冗余 / 错误 / 高风险未覆盖,按 TC 编号列出
When to Use
- 审存量用例资产(祖传用例、他人编写)是否覆盖到位、能否执行
- 审 AI 产出的用例(覆盖 + 可执行性双线)
- 需要一份独立于编写者的审查结论(写时自审不能替代)
When NOT to Use
- 从零写用例 →
test-case-writing - 写用例过程中的自审 →
test-case-writing阶段四(4A/4B 两层审查) - 端到端流水线中的审查环节 → 由
qa调度本 skill,但单用户直接触发本 skill 同样适用 - 代码变更后的回归范围选择 →
regression-testing
工作流
1. 建立可测点基准(分母)
覆盖审查需要一个合法分母,按优先级取:
- 人工标注的可测点清单(存在时,最权威)
- 需求模型 + 与用户共同确认的可测点清单(审查开始前列出,请用户补漏确认)
- 仅有原始输入源 → 从 PRD/代码自行提炼可测点清单,**标注"未经确认"**并在交付时请用户复核
没有分母的覆盖率是给自己批改作业——基准缺失时如实说明,不编造覆盖率。
2. 覆盖审查(此时执行 ../core/coverage.md:核心 7 维逐项全检 + 横切可执行性)
- 核心七维度逐项:功能主流程 / 输入校验 / 逆向操作与生命周期 / 状态流转 / 数据一致性 / 文档隐含需求 / 代码审查发现(有代码时);扩展维度(8–19)按 coverage.md「维度选取速查」按项目类型选取,不做机械全检(执行强度按消费方分流,见其文件头)
- 状态流转维度复核时加载
../core/methods/state-machine.md:按其"状态集 × 事件集 × 转换边"清单逐边核对用例覆盖(每边至少一条 + 非法转换/并发竞态/逆向边三类必补),用例没按状态机组织时反向自行提取状态机再核对,防"看起来有覆盖" - 二阶交叉(
../core/testing-principles.md第 3 节):写入路径 × 校验规则、失败 × 重试、标识 × 重复——存量用例最常见的系统性缺口 - 对照基准逐点核记:已覆盖(TC 编号)/ 未覆盖 / 覆盖但断言错误 / 冗余(多条测同一点)/ 无效(测的不是本需求)
3. 可执行性审查(此时执行 ../core/executability.md 全部检查项)
逐条用例过八条硬标准,重点命中:
- 占位符数据(
{xxx}、"某数据")、虚构入口、模糊判定("功能正常")、异步无时限、断言超强度、前置不可得无 TODO、正文代码内部、缺导读四件套
4. 正确性审查(有 PRD/代码时)
- 用例预期结果与 PRD 规则 / 代码实现是否一致(静态裁决:代码为准,见
../core/evidence.md) - 优先级标注合理性(P0 逐条过自检:失败则核心不可用?);风险等级对齐
../core/risk-model.md(Critical 必有 P0)
5. 修订与落盘
- 直接在用例文件中修订:补缺失用例(追加 TC 编号)、删除冗余、改正错误断言、补可执行性要素(导读区/时限/入口路径/具体数据);新增与改写的用例同样执行
../core/case-format.md格式硬约束(四段式/协作五段式、TC 编号、正文零代码内部) - 修订后重新抽取 Schema(字段与转义规则见
../core/schema-extraction.md),并用../core/scripts/validate_schema.py复验通过后再落盘 - 文件末尾追加审查记录:
## 审查记录(test-case-review {日期})
- 基准:{人工标注 / 需求模型确认 / 自行提炼(未经确认)}
- 审查前:XX 条用例,XX 个模块
- 审查后:XX 条用例,XX 个模块
- 缺失(已补):TC-xx…({场景})
- 冗余(已删):TC-xx…
- 错误(已改):TC-xx…({问题→修正})
- 高风险未覆盖:{风险点 + 建议用例,无代码证据则标注}
- 可执行性修复:{占位符/时限/入口 等 XX 处}
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed bbc137ea241e
- 2mo ago Changed · +14 tokens per session 670978ba849b
- 2mo ago Changed · +1 lines c0bf435130b4
- 2mo ago First seen · 91 lines · 95 tokens per session scan A 2ed62dd30fdb
test-case-review is a skill published in the GitHub repository fishzjp/qa-skills (61 stars, last pushed 3d ago), licensed MIT. It adds 109 tokens to every session and 1,767 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
test-case-review compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| test-case-reviewfishzjp/qa-skills | Skill | 61 | 3d ago | MIT | A | 109 | Claude Code, Codex |
| qa-expertpersonamanagmentlayer/pcl | Skill | 44 | 12d ago | Apache-2.0 | A | 53 | Claude Code |
| defect-analyzerukkuru/testmetry-skills | Skill | 1 | 7mo ago | none | A | 38 | Claude Code |
| test-case-writerukkuru/testmetry-skills | Skill | 1 | 7mo ago | none | A | 50 | Claude Code |
| test-reviewakovalion/paranoid-qa | Skill | 16 | 5d ago | MIT | A | 137 | Claude Code |
| testrail-qauarlouski/testrail-mcp-server | Skill | 44 | yesterday | Apache-2.0 | A | 70 | Claude Code, Codex |
| bug-reportakovalion/paranoid-qa | Skill | 16 | 5d ago | MIT | A | 41 | Claude Code |