← 返回 Skills 市场
105
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install skill-test-skill
功能描述
Tests and scores any Agent Skill against the official anthropics/skills specification. Use this skill when you need to check if a skill repository or SKILL.m...
安全使用建议
This skill is coherent for auditing skills, but be careful: when you run it, only point it at repositories or paths you control. Do NOT point it at system directories, multi-repo workspaces, or any directory that may contain secrets (API keys, private configs, credentials). If you want a lower-risk audit, provide only the SKILL.md or a trimmed copy of the repository (remove secrets and large binaries) or run the auditor in an isolated environment. Consider asking for a mode that omits quoting full file contents or that restricts checks to SKILL.md and references/ files only.
功能分析
Type: OpenClaw Skill
Name: skill-test-skill
Version: 0.1.0
The skill-test-skill is a utility designed to audit and score Agent Skill bundles against the official specification. It provides structured instructions for an AI agent to read skill files (SKILL.md, scripts, references, etc.) and evaluate them across six dimensions to generate a detailed compliance report. The behavior is entirely consistent with its stated purpose, and the provided files (SKILL.md, README.md, and references/) contain no evidence of malicious intent, data exfiltration, or unauthorized command execution. It follows established best practices for skill creation, such as using imperative language and providing clear rationale for its auditing logic.
能力评估
Purpose & Capability
The skill's name/description (audit/score other skills) matches what it asks the agent to do: list a skill directory, fetch GitHub skill directories, and read SKILL.md and auxiliary files for compliance checks. It requests no unrelated binaries, env vars, or installs.
Instruction Scope
SKILL.md instructs the agent to 'read every file' in the provided skill directory (SKILL.md, scripts/, references/, assets/, and any other files) and to quote exact lines/values in the report. This is consistent with an exhaustive auditor but creates a privacy/exfiltration risk if a user points the skill at directories containing secrets or large binary blobs. The instructions do not tell the agent to read files outside the provided path, but they do require fetching GitHub URLs and reading any files found there.
Install Mechanism
No install spec and no code files — instruction-only skill. This is the lowest-risk install profile and matches the skill's description.
Credentials
The skill requires no environment variables, credentials, or config paths. That is proportionate to an auditing skill. Note: because the skill outputs file contents/line quotes, any sensitive values in audited files may be echoed into the report.
Persistence & Privilege
The skill is not always: true and does not request elevated or persistent privileges. The default ability for the agent to invoke the skill autonomously (disable-model-invocation: false) is normal — but combined with the instruction to read arbitrary user-supplied directories, it increases the blast radius if the agent is allowed to call it without explicit user direction.
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-test-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/skill-test-skill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
v0.1.0
Initial release providing full skill audit and scoring against the Agent Skills specification.
- Audits any Agent Skill repository or SKILL.md file for compliance with the official Agent Skills spec.
- Auto-detects the request language (English or Chinese) and generates a detailed compliance report in that language.
- Recursively reads and analyzes all files in the provided skill directory, including scripts, references, and assets.
- Scores the skill across six dimensions: directory structure, frontmatter, content quality, progressive disclosure, optional directory usage, and trigger optimization.
- Outputs a comprehensive Markdown report with findings, compliance score, and prioritized improvement suggestions.
元数据
常见问题
Skill Test Skill 是什么？
Tests and scores any Agent Skill against the official anthropics/skills specification. Use this skill when you need to check if a skill repository or SKILL.m... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 105 次。
如何安装 Skill Test Skill？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install skill-test-skill」即可一键安装，无需额外配置。
Skill Test Skill 是免费的吗？
是的，Skill Test Skill 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Skill Test Skill 支持哪些平台？
Skill Test Skill 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 Test Skill？
由 youngfree（@youngfreefjs）开发并维护，当前版本 v0.1.0。
推荐 Skills