← 返回 Skills 市场
skill-quality-checker
作者
shaozhengmao
· GitHub ↗
· v1.0.0
· MIT-0
256
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install skill-quality-checker
功能描述
Skill 质量检测工具。对已安装的 skills 进行自动化质量评估，按5个维度打分 （问题-方案匹配度、完成度、容错性、Description精度、Token效率），输出评分报告和改进建议。 触发词：检查skill质量、评估skill、skill质量报告、skill review、审查skill、check...
安全使用建议
This tool appears to be a local, static quality scanner that reads skill directories and produces markdown/JSON reports. Before running: (1) inspect scripts/check_quality.py yourself (you already have it bundled); (2) run it against a small test scan-dir first (not your entire skills tree) to see report contents; (3) be aware it will read all files under the scan path, so any accidentally committed secrets in skill files will appear in the output; (4) there are no network calls or credential requests in the visible code, and it does not require installation of external packages. If you want extra assurance, search the full file for imports like subprocess, socket, requests, urllib, or any hard-coded URLs before running.
功能分析
Type: OpenClaw Skill
Name: skill-quality-checker
Version: 1.0.0
The skill is a static analysis utility designed to audit and score other installed OpenClaw skills based on five quality dimensions. The core logic in scripts/check_quality.py uses standard Python libraries to parse markdown and source code for structural patterns (e.g., error handling, file size, metadata) without any evidence of data exfiltration, unauthorized network access, or malicious intent. The instructions in SKILL.md are consistent with the tool's stated purpose of providing quality reports and do not attempt to manipulate the agent into performing unauthorized actions.
能力评估
Purpose & Capability
Name/description (skill-quality-checker) match the included artifacts: a Python script and scoring reference. The script inspects SKILL.md, scripts/, and references/ to compute heuristic scores — this is expected for a quality scanner.
Instruction Scope
Instructions tell the user to run the included Python script against a skills directory (examples use /root/.openclaw/skills/). The script reads and analyzes files under the target skill directories (SKILL.md, scripts, references, etc.). This is coherent for its purpose, but it does mean the tool will read all files under the scan path (so any sensitive data accidentally stored in a skill's files would be included in the report). The script does not appear to execute other skills' scripts or spawn subprocesses (no subprocess usage in the visible imports).
Install Mechanism
No install spec — instruction-only with an included script. No downloads or package installs are required; the script claims to use only the Python standard library, which is consistent with the top-of-file imports.
Credentials
The skill declares no required environment variables, credentials, or config paths and the code does not reference external secrets or request credentials. Nothing disproportionate is requested for a local static analyzer.
Persistence & Privilege
always:false and user-invocable; the skill does not request persistent or elevated platform privileges. It reads files in the provided scan directory but does not modify other skills or system configuration.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install skill-quality-checker - 安装完成后，直接呼叫该 Skill 的名称或使用
/skill-quality-checker触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Skill Quality Inspector (Initial Release)
An automated tool designed to evaluate the quality of installed skills and generate comprehensive scoring reports with optimization suggestions.
Multi-Dimensional Assessment: Supports batch or individual skill checks across 5 core metrics: Intent Fit, Task Completeness, Robustness, Description Accuracy, and Token Efficiency.
Flexible Output: Supports Markdown and JSON formats for easy viewing and archiving.
Zero Dependencies: Performs full-process static analysis using only the Python Standard Library.
Comprehensive Documentation: Includes detailed usage guides and scoring criteria references.
Robust Workflow: Features clear grading scales and graceful error handling (e.g., proactive alerts for missing directories).
元数据
常见问题
skill-quality-checker 是什么？
Skill 质量检测工具。对已安装的 skills 进行自动化质量评估，按5个维度打分 （问题-方案匹配度、完成度、容错性、Description精度、Token效率），输出评分报告和改进建议。 触发词：检查skill质量、评估skill、skill质量报告、skill review、审查skill、check... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 256 次。
如何安装 skill-quality-checker？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install skill-quality-checker」即可一键安装，无需额外配置。
skill-quality-checker 是免费的吗？
是的，skill-quality-checker 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
skill-quality-checker 支持哪些平台？
skill-quality-checker 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 skill-quality-checker？
由 shaozhengmao（@shaozhengmao）开发并维护，当前版本 v1.0.0。
推荐 Skills