Agent审查
- 作者仓库星标 386
- 作者更新于 实时读取
- 作者仓库 skillport
- 领域
- 安全
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @gotalab · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 即装即用
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- Windows
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生，可能与作者实际意图存在差异。
需要注意： 未限定 allowed-tools，默认拥有全部工具权限。
---
name: skill-evaluator
description: Evaluates agent skills against Anthropic's best practices. Use when asked to review, evaluate, a…
category: 安全
runtime: 无特殊运行时
---
# skill-evaluator 输出预览
## PART A: 任务判断
- 适用问题：安全审计、密钥扫描、权限检查或风险分析。
- 输入要求：目标材料、限制条件、期望输出和验收方式。
- 证据边界：围绕“Quick Start / Evaluation Workflow / Step 1: Automated Validation”读取原文规则，不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断：确认你的需求是否属于安全审计、密钥扫描、权限检查或风险分析，并标出输入、限制和预期结果。
- **02** 执行计划：优先按“Quick Start / Evaluation Workflow / Step 1: Automated Validation”拆成步骤，说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果：给出可复制的命令、文件改动、检查清单或内容草稿，并说明如何继续迭代。
- **04** 风险边界：结合 读取文件、写入/修改文件、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件；主要在本地完成；通常不需要额外 API Key。
- 先小样例验证，再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 先确认触发方式
原文没有稳定的斜杠命令要求。安装验证后通常全局生效，直接在对话里点名这个 Skill 并描述任务即可。
给清楚输入和边界
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是：读取文件、写入/修改文件。
小样例验证后再放大
先用一个小任务确认它会围绕“Quick Start / Evaluation Workflow / Step 1: Automated Validation”工作；涉及文件或命令时，先看 diff、日志、预览或测试结果。
复核后再交付
检查最终产物是否包含明确结果、必要证据和下一步动作；如果输出泛泛而谈，就补充输入、边界和验收标准后重跑。
---
name: skill-evaluator
description: Evaluates agent skills against Anthropic's best practices. Use when asked to review, evaluate, a…
category: 安全
source: gotalab/skillport
---
# skill-evaluator
## 什么时候使用
- 用于审阅代码、文档或方案并给出可执行反馈 适合处理安全审计、密钥扫描、权限检查和风险分析，核心价值是把输入、判断、执行、验证和交付边界固定下来，避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代的步骤；通常不需要额外 A…
- 面向安全审计、密钥扫描、权限检查或风险分析，优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「Quick Start / Evaluation Workflow / Step 1: Automated Validation」组织步骤，不把推断写成作者事实。
- 读取文件、写入/修改文件；主要在本地完成；通常不需要额外 API Key。
- 先跑小样例，确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown，不编造命令、平台或依赖。 证据边界与执行链路
作者原文负责流程事实；仓库文件负责来源和命令；流狐只补充适用场景、限制和质量判断。
skill "skill-evaluator" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> Quick Start / Evaluation Workflow / Step 1: Automated Validation
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
}Skill Evaluator (WIP)
Evaluates skills against Anthropic's official best practices for agent skill authoring. Produces structured evaluation reports with scores and actionable recommendations.
Quick Start
- Read the skill's SKILL.md and understand its purpose
- Run automated validation:
scripts/validate_skill.py <skill-path> - Perform manual evaluation against criteria below
- Generate evaluation report with scores and recommendations
Evaluation Workflow
Step 1: Automated Validation
Run the validation script first:
scripts/validate_skill.py <path/to/skill>This checks:
- SKILL.md exists with valid YAML frontmatter
- Name follows conventions (lowercase, hyphens, max 64 chars)
- Description is present and under 1024 chars
- Body is under 500 lines
- File references are one-level deep
Step 2: Manual Evaluation
Evaluate each dimension and assign a score (1-5):
A. Naming (Weight: 10%)
| Score | Criteria |
|---|---|
| 5 | Gerund form (-ing), clear purpose, memorable |
| 4 | Descriptive, follows conventions |
| 3 | Acceptable but could be clearer |
| 2 | Vague or misleading |
| 1 | Violates naming rules |
Rules: Max 64 chars, lowercase + numbers + hyphens only, no reserved words (anthropic, claude), no XML tags.
Good: processing-pdfs, analyzing-spreadsheets, building-dashboards
Bad: pdf, my-skill, ClaudeHelper, anthropic-tools
B. Description (Weight: 20%)
| Score | Criteria |
|---|---|
| 5 | Clear functionality + specific activation triggers + third person |
| 4 | Good description with some triggers |
| 3 | Adequate but missing triggers or vague |
| 2 | Too brief or unclear purpose |
| 1 | Missing or unhelpful |
Must include: What the skill does AND when to use it. Good: "Extracts text from PDFs. Use when working with PDF documents for text extraction, form parsing, or content analysis." Bad: "A skill for PDFs." or "Helps with documents."
C. Content Quality (Weight: 30%)
| Score | Criteria |
|---|---|
| 5 | Concise, assumes Claude intelligence, actionable instructions |
| 4 | Generally good, minor verbosity |
| 3 | Some unnecessary explanations or redundancy |
| 2 | Overly verbose or confusing |
| 1 | Bloated, explains obvious concepts |
Ask: "Does Claude really need this explanation?" Remove anything Claude already knows.
D. Structure & Organization (Weight: 25%)
| Score | Criteria |
|---|---|
| 5 | Excellent progressive disclosure, clear navigation, optimal length |
| 4 | Good organization, appropriate file splits |
| 3 | Acceptable but could be better organized |
| 2 | Poor organization, missing references, or bloated SKILL.md |
| 1 | No structure, everything dumped in SKILL.md |
Check:
- SKILL.md under 500 lines
- References are one-level deep (no nested chains)
- Long reference files (>100 lines) have table of contents
- Uses forward slashes in all paths
E. Degrees of Freedom (Weight: 10%)
| Score | Criteria |
|---|---|
| 5 | Perfect match: high freedom for flexible tasks, low for fragile operations |
| 4 | Generally appropriate freedom levels |
| 3 | Acceptable but could be better calibrated |
| 2 | Mismatched: too rigid or too loose |
| 1 | Completely wrong freedom level for the task type |
Guideline:
- High freedom (text): Multiple valid approaches, context-dependent
- Medium freedom (parameterized): Preferred pattern exists, some variation OK
- Low freedom (specific scripts): Fragile operations, exact sequence required
F. Anti-Pattern Check (Weight: 5%)
Deduct points for each anti-pattern found:
- Too many options without clear recommendation (-1)
- Time-sensitive information with date conditionals (-1)
- Inconsistent terminology (-1)
- Windows-style paths (backslashes) (-1)
- Deeply nested references (more than one level) (-2)
- Scripts that punt error handling to Claude (-1)
- Magic numbers without justification (-1)
Step 3: Generate Report
Use this template:
# Skill Evaluation Report: [skill-name]
## Summary
- **Overall Score**: X.X/5.0
- **Recommendation**: [Ready for publication / Needs minor improvements / Needs major revision]
## Dimension Scores
| Dimension | Score | Weight | Weighted |
|-----------|-------|--------|----------|
| Naming | X/5 | 10% | X.XX |
| Description | X/5 | 20% | X.XX |
| Content Quality | X/5 | 30% | X.XX |
| Structure | X/5 | 25% | X.XX |
| Degrees of Freedom | X/5 | 10% | X.XX |
| Anti-Patterns | X/5 | 5% | X.XX |
| **Total** | | 100% | **X.XX** |
## Strengths
- [List 2-3 things done well]
## Areas for Improvement
- [List specific issues with actionable fixes]
## Anti-Patterns Found
- [List any anti-patterns detected]
## Recommendations
1. [Priority 1 fix]
2. [Priority 2 fix]
3. [Priority 3 fix]
## Pre-Publication Checklist
- [ ] Description is specific with activation triggers
- [ ] SKILL.md under 500 lines
- [ ] One-level-deep file references
- [ ] Forward slashes in all paths
- [ ] No time-sensitive information
- [ ] Consistent terminology
- [ ] Concrete examples provided
- [ ] Scripts handle errors explicitly
- [ ] All configuration values justified
- [ ] Required packages listed
- [ ] Tested with Haiku, Sonnet, OpusScore Interpretation
| Score Range | Rating | Action |
|---|---|---|
| 4.5 - 5.0 | Excellent | Ready for publication |
| 4.0 - 4.4 | Good | Minor improvements recommended |
| 3.0 - 3.9 | Acceptable | Several improvements needed |
| 2.0 - 2.9 | Needs Work | Major revision required |
| 1.0 - 1.9 | Poor | Fundamental redesign needed |
References
- references/evaluation-criteria.md - Detailed evaluation criteria with examples
- references/scoring-rubric.md - Complete scoring rubric and edge cases
Examples
See evaluations/ for example evaluation scenarios.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核