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 Kokxi/qa-test-skills --skill qa-agent-testinggit clone --depth 1 https://github.com/Kokxi/qa-test-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/kokxi/qa-test-skills/qa-agent-testing)<a href="https://agentmods.dev/skills/kokxi/qa-test-skills/qa-agent-testing"><img src="https://agentmods.dev/badge/skills/kokxi/qa-test-skills/qa-agent-testing.svg" alt="Measured on agentmods" 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 | $0.00119 | $0.02834 |
| Opus 5 | $0.00060 | $0.01417 |
| Sonnet 5 | $0.00024 | $0.00567 |
| Haiku 4.5 | $0.00012 | $0.00283 |
Grade A, and why
qa-agent-testing 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 2d 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 — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Agent测试专项
核心原则
Agent测试的核心——验证AI决策的正确性、安全性、可控性。
启动方式：用户提出Agent测试需求后，按Agent类型速查表定位必测维度，输出测试方案。
深度要求
| 复杂度 | 用例数要求 | 说明 |
|---|---|---|
| 简单Agent | 30条 | 单一任务Agent |
| 中等Agent | 50条 | 多任务Agent |
| 复杂Agent | 80条 | 多工具/多轮对话Agent |
必须覆盖的9个维度：
| 维度 | 占比 | 说明 |
|---|---|---|
| 功能测试 | 25% | 任务执行/决策/交互/工具调用 |
| 安全测试 | 15% | Prompt注入/越权/敏感信息 |
| 高级安全测试 | 12% | 间接注入/多轮诱导/编码绕过 |
| 边界测试 | 10% | 输入/能力/并发边界 |
| 可控性 | 12% | 中止/人工确认/权限边界/速率限制 |
| 可靠性测试 | 8% | 稳定性/容错/降级 |
| 幻觉与事实性 | 8% | 事实核查/来源归因/RAG准确性 |
| 推理链路 | 5% | 可解释性/逻辑/自纠错 |
| 工具调用测试 | 5% | 参数生成/工具链编排/副作用 |
每个维度的详细测试范围、典型用例和检查清单参见
references/test-framework.md。
Agent类型速查
不同Agent类型各有侧重：
| Agent类型 | 典型代表 | 必测维度 | 重点关注 |
|---|---|---|---|
| 对话助手型 | AI客服/智能导购/知识问答 | 功能+安全+幻觉 | 意图识别、上下文记忆、幻觉控制、对话流畅度 |
| 任务执行型 | 工单处理/审批流转/数据录入 | 功能+工具调用+可控性 | 工具选择、参数生成、执行顺序、人工确认 |
| 数据分析型 | BI助手/报表生成/趋势分析 | 功能+幻觉+推理 | 数据准确性、来源归因、逻辑正确性、图表输出 |
| 自主决策型 | 风控系统/资源调度/智能运维 | 安全+可控性+推理+工具调用 | 间接注入、权限边界、HITL、决策归因 |
测试方案输出结构
AI加载此技能后输出的测试方案：
1. Agent类型识别 → 判断属于哪一类（对话/任务/分析/决策），列出判定理由
2. 必测维度清单 → 从9维中筛选该类型必须覆盖的维度
3. 测试范围详解 → 每个必测维度的核心测试点
4. 典型用例参考 → 从46条典型用例中选取适用的
5. 安全与可控性专项 → 注入/工具安全/HITL等Agent特有风险验证
6. 风险提示 → 基于场景的高风险区域预警
测试用例设计
用例模板
## Agent测试用例
### 基本信息
- 用例编号：AGENT-XXX
- 测试类型：功能/安全/高级安全/边界/可控性/可靠性/幻觉/推理/工具调用
- 测试目标：[具体目标]
### 测试场景
- 输入：[用户输入/指令]
- 上下文：[历史对话/环境信息]
- 期望行为：[Agent应该如何响应]
### 测试步骤
1. [步骤1]
2. [步骤2]
3. [步骤3]
### 预期结果
- 行为：[Agent的行为]
- 输出：[Agent的输出]
- 安全：[安全检查结果]
### 风险等级
高/中/低
46条各维度的典型用例参见
references/test-framework.md的"典型用例"章节。
输出示例
用户说"帮我测试这个AI客服Agent" → 启动九维测试：功能（对话/意图/上下文）→ 安全（注入/越权/敏感）→ 高级安全（间接注入/多轮诱导/编码绕过）→ 边界（空/超长/并发）→ 可控性（HITL/中止/权限）→ 可靠性（长时间/降级）→ 幻觉（事实核查/来源归因/RAG）→ 推理链路（逻辑/自纠错）→ 工具调用（选择/参数/编排） → 输出Agent测试方案
Agent出现幻觉回答 → 启动幻觉与事实性检查，核查信息真实性和来源归因，同时排查是否由注入导致
场景：测试工单处理Agent（任务执行型） → 功能（全流程正确）→ 工具调用（参数完整/顺序正确）→ 可控性（确认节点/取消机制）→ 边界（100并发） 输出：按功能→工具→可控性→边界优先级的测试方案
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago Changed e891d3968cf8
- 7d ago First seen · 201 lines · 119 tokens per session scan A 48691bf35824
qa-agent-testing is a skill published in the GitHub repository Kokxi/qa-test-skills (24 stars, last pushed 5d ago), licensed MIT. It adds 119 tokens to every session and 2,834 once invoked, about $0.0006 per session on Opus 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.
Other skills, from other repositories
agent-device
Automates Apple-platform apps (iOS, tvOS, macOS), Android devices, and Amazon Vega OS TV apps in Vega Virtual Devices. Use when navigating apps, taking snapshots/screenshots where supported, driving TV remotes, tapping, typing, scrolling, extracting UI info, collecting evidence, or planning agent-device CLI commands.
android-emulator
Verify and debug native, React Native, Expo, or Flutter apps on an Android Emulator with agent-device. Use when an agent needs to launch an app, inspect its live UI, tap, type, scroll, validate a code change, collect failure evidence, or reproduce a workflow on an Android virtual device.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
ios-simulator
Verify and debug native, React Native, Expo, or Flutter apps on an iOS Simulator with agent-device. Use when an agent needs to launch an app, inspect its live UI, tap, type, scroll, validate a code change, collect failure evidence, or reproduce a workflow on an iPhone or iPad Simulator.
verification-engine
Use when verifying build/test/lint before commit, PR, or completion claims. Runs verification pipeline in fresh subagent context with auto-repair. Triggers on /handoff-verify, pre-commit check, build verification, test validation.
verify-implementation
A workflow that runs a project’s verification skills to produce a report on coding patterns, architecture rules, and project conventions. It is intended for work after implementation, before a pull request, or during code review.