← Back to Skills Marketplace
Skill Quality Assurance
by
pagoda111king
· GitHub ↗
· v0.1.0
· MIT-0
84
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install skill-quality-assurance
Description
技能质量六维评估助手。使用结构化六维评估模型（T 技术深度、C 认知增强、O 编排能力、E 进化能力、M 市场验证、U 用户体验）对技能进行自动化评估。使用当：(1) 新技能开发完成后需要质量评估，(2) 现有技能需要版本升级评估，(3) 准备 ClawHub 上架前需要质量检查，(4) 定期技能健康检查。
Usage Guidance
该技能自成一体，功能是本地静态评估技能包并生成 Markdown 报告。建议在安装/运行前：1) 在沙箱或非敏感目录下运行首次评估以确认行为（因为评估器会读取传入的 skillPath）；2) 如果允许代理自动调用技能，考虑信任边界——不要让代理传入系统根目录或包含密钥/凭证的路径；3) 如需更高保障，可人工审计 scripts/evaluation-engine.js（目前仅使用 fs 读取并打分，未见网络请求或子进程调用）；4) 若你希望该技能评估远端仓库或自动下载代码，先确认网络/凭证流程并审查相应实现。
Capability Analysis
Type: OpenClaw Skill
Name: skill-quality-assurance
Version: 0.1.0
The skill bundle is a quality assurance tool designed to evaluate other OpenClaw skills using a structured six-dimensional model (TCOEMU). The core logic in 'scripts/evaluation-engine.js' performs heuristic analysis by reading local skill files and checking for the presence of specific directories and keywords to generate a Markdown report. There is no evidence of malicious intent, data exfiltration, or unauthorized execution; the script uses standard Node.js 'fs' and 'path' modules without external network calls or shell commands.
Capability Assessment
Purpose & Capability
名称和描述是“技能质量评估”，所包含的文件（SKILL.md、references、assets、scripts/evaluation-engine.js）以及运行时行为（读取技能路径并生成报告）与其目的匹配。没有声明与任务无关的 env/凭证或二进制依赖。
Instruction Scope
SKILL.md 指示读取目标技能的 SKILL.md、目录结构、测试、文档等；scripts/evaluation-engine.js 实现也只是同步读取目录和文件并打分。注意：评估引擎按传入的 skillPath 读取任意路径，若被指向系统或敏感目录会暴露本地文件内容——这是功能范围内但需注意的权限边界。
Install Mechanism
无安装规范（instruction-only + 附带脚本）。没有从不受信任的 URL 下载、没有在安装阶段写入或执行外部二进制。
Credentials
不要求任何环境变量或凭证（requires.env 为空），脚本也没有引用外部 API 密钥或托管凭证，所需权限与评估任务匹配。
Persistence & Privilege
flags 中没有 always:true，默认的可被用户或代理调用的权限与常规技能一致。脚本不修改其他技能配置或全局 agent 设置。
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install skill-quality-assurance - After installation, invoke the skill by name or use
/skill-quality-assurance - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
Initial release: provides an automated six-dimension quality evaluation tool for skills.
- Implements structured T/C/O/E/M/U dimensional scoring and reporting
- Generates actionable evaluation reports and prioritized improvement suggestions (P0/P1/P2)
- Includes pre-listing quality checklist for ClawHub
- Adopts design patterns: Evaluation Factory, Progressive Disclosure, and Six-Dimensions Model
- Supports both single and batch skill evaluation via CLI
Metadata
Frequently Asked Questions
What is Skill Quality Assurance?
技能质量六维评估助手。使用结构化六维评估模型（T 技术深度、C 认知增强、O 编排能力、E 进化能力、M 市场验证、U 用户体验）对技能进行自动化评估。使用当：(1) 新技能开发完成后需要质量评估，(2) 现有技能需要版本升级评估，(3) 准备 ClawHub 上架前需要质量检查，(4) 定期技能健康检查。 It is an AI Agent Skill for Claude Code / OpenClaw, with 84 downloads so far.
How do I install Skill Quality Assurance?
Run "/install skill-quality-assurance" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Skill Quality Assurance free?
Yes, Skill Quality Assurance is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Skill Quality Assurance support?
Skill Quality Assurance is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Skill Quality Assurance?
It is built and maintained by pagoda111king (@pagoda111king); the current version is v0.1.0.
More Skills