功能描述

广州日报/融媒云通用测试助手。专为广州日报报业集团及广州市区融媒云平台的测试工作设计。覆盖场景： (1) 广州市区融媒云平台后台管理系统（CMS/采编/运营后台）功能测试 (2) 广州融媒云APP（Android/iOS/HarmonyOS）功能、性能、兼容性测试 (3) 广州日报新花城APP（Android/...

使用说明 (SKILL.md)

广州日报/融媒云通用测试助手

Name: 广州日报/融媒云通用测试助手
Author: freeeye

平台概览

平台	说明
融媒云后台	广州市区融媒体云平台采编/运营管理后台（B/S架构，内网+公网双通道）
融媒云APP	广州日报社开发，包ID: `com.pdmi.blend.media`，Android/iOS/HarmonyOS三端
新花城APP	广州日报官方客户端，广州市区融媒体平台，Android/iOS/HarmonyOS三端
新花城H5	域名: `huacheng.gz-cmc.com`，内嵌于APP及微信生态
大洋网	`news.dayoo.com`，广州日报新闻门户
微信小程序	新花城小程序、各区融媒体小程序

工作流

1. 接收测试任务

用户说明测试目标后，先确认：

测试类型：冒烟 / 功能 / 回归 / 性能 / 接口 / 兼容性
测试平台：后台 / APP（哪个端）/ H5 / 小程序 / 接口
测试范围：模块/功能点/版本号
输出格式：测试用例 / 测试报告 / Bug单

如信息不足，主动询问缺失项（一次最多问3个问题）。

2. 生成测试用例

调用 scripts/generate_cases.py 或直接按模板输出。参见 references/test-cases.md 获取各平台标准用例库。

3. 执行接口测试

调用 scripts/api_test.py 对目标接口发起请求并验证响应。接口规范参见 references/api-reference.md。

4. 输出测试报告

按 references/report-template.md 格式输出测试报告。

5. 撰写 Bug 单

按 references/bug-template.md 格式输出 Bug 单。

快速命令

用户说	执行动作
"生成冒烟测试用例"	读 references/test-cases.md → 输出对应平台冒烟用例
"测试这个接口 [URL]"	运行 scripts/api_test.py
"生成测试报告"	读 references/report-template.md → 填充输出
"写一个Bug单"	读 references/bug-template.md → 填充输出
"检查H5链接"	运行 scripts/check_url.py
"APP兼容性测试方案"	读 references/compatibility.md

注意事项

接口测试默认使用测试环境，不得对生产环境发起写操作
APP测试需区分 Android / iOS / HarmonyOS 三端差异
H5测试需覆盖微信内置浏览器、系统浏览器、APP内嵌WebView三种场景
性能基准：APP冷启动 ≤ 3s，接口响应 ≤ 1s（P99），H5首屏 ≤ 2s

安全使用建议

This skill appears coherent for QA/testing of the Guangzhou media platforms. Before installing/using it: 1) Run it in a sandbox or isolated environment because the scripts will perform arbitrary HTTP requests and may auto-install the 'requests' package via pip. 2) Do not point the security or performance tests at production systems — the api_test tool includes SQLi/XSS and load tests that can be disruptive. 3) If you plan to supply authentication tokens, treat them as sensitive: pass them only to intended test environments and avoid storing them in shared locations. 4) Review the preset target URLs and change them if you don't want tests to probe those hosts. 5) If you have stricter change controls, consider extracting the three scripts and running them under your own supervision rather than allowing the agent to invoke them autonomously.

功能分析

Type: OpenClaw Skill Name: gz-media-tester Version: 1.0.0 The skill bundle is a legitimate testing suite designed for the Guangzhou Daily and Rongmei Cloud media platforms. It contains Python scripts for API testing (api_test.py), URL availability checking (check_url.py), and test case generation (generate_cases.py), along with comprehensive documentation and templates. While the scripts include functionality to automatically install the 'requests' library via pip and perform network requests (including simulated SQL injection and XSS payloads for target verification), these behaviors are strictly aligned with the stated purpose of a QA and security testing tool. No evidence of data exfiltration, malicious persistence, or harmful prompt injection was found.

能力评估

✓ Purpose & Capability

Name/description (platform/App/H5/API testing) match the included reference docs and three scripts (generate_cases.py, api_test.py, check_url.py). Required env/config/credentials are none, which aligns with a general testing helper that accepts target URLs/tokens at runtime.

ℹ Instruction Scope

SKILL.md instructs the agent to generate cases, run API tests, check URLs, and produce reports — exactly what the bundled scripts implement. The api_test tool explicitly includes security tests (SQLi/XSS) and a performance test that issues many requests; this is expected for a QA tool but could be destructive if pointed at production systems. The skill documentation does state to avoid write operations against production, which mitigates but does not eliminate operational risk.

ℹ Install Mechanism

There is no registry install spec (instruction-only), but both runtime scripts auto-install the 'requests' package via subprocess pip install if missing. That means the skill may modify the Python environment and requires outbound network access during the first run — expected but worth noting.

✓ Credentials

The skill declares no required environment variables or credentials. The scripts accept tokens/headers as CLI args for authenticated API tests; this is proportionate to the stated purpose. There are no requests for unrelated secrets or config paths.

✓ Persistence & Privilege

Skill is not always-enabled and does not declare elevated persistence. It does not modify other skills' configuration. Runtime behavior is limited to running the included scripts and network calls to targets provided by the user.

版本历史

v1.0.0

- 初始发布，推出广州日报/融媒云统一测试助手。 - 支持后台、APP、H5、小程序、后端接口等多种测试场景。 - 可自动生成测试用例、测试报告和Bug单。 - 添加测试类型、平台、范围自动识别及补全询问流程。 - 提供快速命令及标准模板引用，提升测试效率。

元数据

Slug gz-media-tester

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题