@chongyan/autospec 1.0.2 → 1.0.4

package/plugins/benchmarks/templates/skills/project-evolve-learner-template.yaml

@@ -1,50 +0,0 @@
-# AutoSpec Skill Benchmark Template - Project-Evolve-Learner
-# 适用于: 测试 project-evolve-learner skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-project-evolve-learner"
-description: "Project-Evolve-Learner Skill 基准测试"
-
-type: skill
-target: project-evolve-learner
-
-testCases:
-  - name: "pattern-learning"
-    input:
-      context: "从实践日志中学习模式"
-    complexity: 3
-    expectedBehaviors:
-      - "读取 practice-log"
-      - "识别重复模式"
-      - "提取通用解决方案"
-    expectedOutput:
-      - "模式识别报告"
-      - "通用解决方案"
-    successCriteria:
-      - "模式识别准确"
-      - "解决方案合理"
-    qualityMetrics:
-      - "模式识别率 >= 80%"
-    maxDuration: 300
-
-  - name: "knowledge-update"
-    input:
-      context: "更新项目知识"
-    complexity: 3
-    expectedBehaviors:
-      - "分析新发现"
-      - "更新知识文件"
-      - "验证更新效果"
-    expectedOutput:
-      - "更新后的知识文件"
-      - "更新报告"
-    successCriteria:
-      - "知识更新正确"
-    qualityMetrics:
-      - "更新准确率 >= 90%"
-    maxDuration: 300
-
-successCriteria:
-  passRate: 80
-  avgFieldCompletion: 85

package/plugins/benchmarks/templates/skills/project-evolve-reviewer-template.yaml

@@ -1,50 +0,0 @@
-# AutoSpec Skill Benchmark Template - Project-Evolve-Reviewer
-# 适用于: 测试 project-evolve-reviewer skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-project-evolve-reviewer"
-description: "Project-Evolve-Reviewer Skill 基准测试"
-
-type: skill
-target: project-evolve-reviewer
-
-testCases:
-  - name: "code-review"
-    input:
-      context: "审查代码变更"
-    complexity: 3
-    expectedBehaviors:
-      - "读取变更内容"
-      - "执行代码审查"
-      - "识别问题"
-      - "生成审查报告"
-    expectedOutput:
-      - "审查报告"
-      - "问题清单"
-    successCriteria:
-      - "审查维度完整"
-      - "问题识别准确"
-    qualityMetrics:
-      - "问题检出率 >= 85%"
-    maxDuration: 300
-
-  - name: "design-review"
-    input:
-      context: "审查设计文档"
-    complexity: 3
-    expectedBehaviors:
-      - "读取设计文档"
-      - "执行设计审查"
-      - "检查一致性"
-    expectedOutput:
-      - "设计审查报告"
-    successCriteria:
-      - "设计审查完整"
-    qualityMetrics:
-      - "一致性检查执行率 = 100%"
-    maxDuration: 300
-
-successCriteria:
-  passRate: 85
-  avgFieldCompletion: 90

package/plugins/benchmarks/templates/skills/project-evolve-scanner-template.yaml

@@ -1,75 +0,0 @@
-# AutoSpec Skill Benchmark Template - Project-Evolve-Scanner
-# 适用于: 测试 project-evolve-scanner skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-project-evolve-scanner"
-description: "Project-Evolve-Scanner Skill 基准测试"
-
-type: skill
-target: project-evolve-scanner
-
-testCases:
-  - name: "nodejs-project-scan"
-    input:
-      context: "扫描一个 Node.js + React 项目"
-    complexity: 1
-    expectedBehaviors:
-      - "检测 package.json"
-      - "识别技术栈"
-      - "分析目录结构"
-      - "发现评测点"
-    expectedOutput:
-      - "项目特征报告"
-      - "技术栈列表"
-      - "评测点列表"
-    successCriteria:
-      - "技术栈检测正确"
-      - "评测点发现完整"
-    qualityMetrics:
-      - "技术栈识别准确率 >= 95%"
-      - "评测点发现率 >= 80%"
-    maxDuration: 300
-
-  - name: "java-project-scan"
-    input:
-      context: "扫描一个 Java Spring Boot 项目"
-    complexity: 3
-    expectedBehaviors:
-      - "检测 pom.xml 或 build.gradle"
-      - "识别技术栈"
-      - "分析依赖版本"
-    expectedOutput:
-      - "项目特征报告"
-      - "依赖分析报告"
-    successCriteria:
-      - "技术栈检测正确"
-      - "依赖分析完整"
-    qualityMetrics:
-      - "技术栈识别准确率 >= 95%"
-      - "依赖分析完整率 >= 90%"
-    maxDuration: 450
-
-  - name: "multi-system-scan"
-    input:
-      context: "扫描包含前后端的多子系统项目"
-    complexity: 5
-    expectedBehaviors:
-      - "检测多个子系统"
-      - "识别各子系统技术栈"
-      - "分析子系统依赖关系"
-      - "生成记忆结构"
-    expectedOutput:
-      - "多系统特征报告"
-      - "系统间依赖关系图"
-    successCriteria:
-      - "子系统识别完整"
-      - "依赖关系清晰"
-    qualityMetrics:
-      - "子系统识别率 = 100%"
-      - "技术栈识别准确率 >= 95%"
-    maxDuration: 600
-
-successCriteria:
-  passRate: 85
-  avgFieldCompletion: 90

package/plugins/benchmarks/templates/skills/project-evolve-template.yaml

@@ -1,72 +0,0 @@
-# AutoSpec Skill Benchmark Template - Project-Evolve
-# 适用于: 测试 project-evolve skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-project-evolve"
-description: "Project-Evolve Skill 基准测试 - AI Native 项目自进化"
-
-type: skill
-target: project-evolve
-
-testCases:
-  - name: "init-memory"
-    input:
-      context: "初始化项目记忆"
-    complexity: 1
-    expectedBehaviors:
-      - "扫描项目技术栈"
-      - "检测项目结构"
-      - "自动发现评测点"
-      - "创建记忆目录结构"
-    expectedOutput:
-      - ".autospec/memory/index.yaml"
-      - "初始化报告"
-    successCriteria:
-      - "技术栈识别正确"
-      - "评测点发现 >= 10 个"
-    qualityMetrics:
-      - "记忆创建率 = 100%"
-    maxDuration: 300
-
-  - name: "cruise-mode"
-    input:
-      context: "智能巡航模式"
-    complexity: 3
-    expectedBehaviors:
-      - "加载 validated 记忆"
-      - "执行代码扫描"
-      - "智能优先级排序"
-      - "L1 自动修复"
-    expectedOutput:
-      - "cruise-report.md"
-      - "auto-fixes.json"
-    successCriteria:
-      - "巡航执行完整"
-      - "自动修复正确"
-    qualityMetrics:
-      - "自动修复准确率 >= 90%"
-    maxDuration: 600
-
-  - name: "deep-evolution"
-    input:
-      context: "深度进化模式"
-    complexity: 5
-    expectedBehaviors:
-      - "六层全维度扫描"
-      - "关联分析"
-      - "生成改进方案"
-      - "区分 auto-fixable/manual-review"
-    expectedOutput:
-      - "deep-report.md"
-      - "improvements.md"
-    successCriteria:
-      - "六层扫描完整"
-      - "改进方案可执行"
-    qualityMetrics:
-      - "维度覆盖率 = 100%"
-    maxDuration: 1200
-
-successCriteria:
-  passRate: 85
-  avgFieldCompletion: 90

package/plugins/benchmarks/templates/skills/project-evolve-verifier-template.yaml

@@ -1,51 +0,0 @@
-# AutoSpec Skill Benchmark Template - Project-Evolve-Verifier
-# 适用于: 测试 project-evolve-verifier skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-project-evolve-verifier"
-description: "Project-Evolve-Verifier Skill 基准测试"
-
-type: skill
-target: project-evolve-verifier
-
-testCases:
-  - name: "fix-verification"
-    input:
-      context: "验证修复效果"
-    complexity: 3
-    expectedBehaviors:
-      - "重新执行失败测试"
-      - "对比修复前后"
-      - "检测退化"
-      - "计算质量分数"
-    expectedOutput:
-      - "验证报告"
-      - "质量分数对比"
-    successCriteria:
-      - "验证执行完整"
-      - "退化检测正确"
-    qualityMetrics:
-      - "验证完整率 = 100%"
-      - "退化检出率 = 100%"
-    maxDuration: 450
-
-  - name: "regression-check"
-    input:
-      context: "回归检测"
-    complexity: 3
-    expectedBehaviors:
-      - "执行全量测试"
-      - "对比历史结果"
-      - "识别退化"
-    expectedOutput:
-      - "回归报告"
-    successCriteria:
-      - "回归检测完整"
-    qualityMetrics:
-      - "退化检出率 = 100%"
-    maxDuration: 600
-
-successCriteria:
-  passRate: 90
-  avgFieldCompletion: 95

package/plugins/benchmarks/templates/skills/requirement-analyzer-template.yaml

@@ -1,48 +0,0 @@
-# AutoSpec Skill Benchmark Template - Requirement Analyzer
-# 适用于: 测试 requirement-stage skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-requirement-analyzer"
-description: "Requirement Analyzer Skill 基准测试"
-
-type: skill
-target: requirement-stage
-
-testCases:
-  - name: "basic-requirement"
-    input:
-      context: "实现用户登录功能"
-      clarifications:
-        - "支持用户名密码登录"
-        - "需要记住密码功能"
-        - "支持手机号验证码登录"
-    expected:
-      artifacts:
-        - "requirement.md"
-      quality:
-        fieldCompletionRate: 90
-        hasAcceptanceCriteria: true
-        hasBoundaryConditions: true
-    maxDuration: 600
-
-  - name: "complex-requirement"
-    input:
-      context: "实现电商订单系统"
-      clarifications:
-        - "支持多种支付方式"
-        - "需要库存扣减"
-        - "支持订单取消和退款"
-    expected:
-      artifacts:
-        - "requirement.md"
-      quality:
-        fieldCompletionRate: 95
-        hasAcceptanceCriteria: true
-        hasBoundaryConditions: true
-        hasAssumptions: true
-    maxDuration: 900
-
-successCriteria:
-  passRate: 85
-  avgFieldCompletion: 90

package/plugins/benchmarks/templates/skills/skill-forge-template.yaml

@@ -1,117 +0,0 @@
-# AutoSpec Skill Benchmark Template - Skill-Forge
-# 适用于: 测试 skill-forge skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-skill-forge"
-description: "Skill-Forge Skill 基准测试 - 技能锻造系统"
-
-type: skill
-target: skill-forge
-
-testCases:
-  - name: "distill-new-skill"
-    input:
-      context: "practice-log 中记录了 5 次模式"
-    complexity: 3
-    expectedBehaviors:
-      - "收集至少 3 条支撑证据"
-      - "执行流程蒸馏"
-      - "生成符合模板的 skill 草稿"
-      - "包含 CSO description 和反模式清单"
-      - "执行 Layer1+Layer2 验证"
-    expectedOutput:
-      - "新 skill 文件"
-      - "锻造报告"
-      - "证据清单"
-    successCriteria:
-      - "证据收集 >= 3 条"
-      - "skill 符合模板标准"
-      - "反模式清单 >= 5 条"
-    qualityMetrics:
-      - "证据收集完整率 = 100%"
-      - "skill 规范率 >= 90%"
-    maxDuration: 600
-
-  - name: "iterate-skill"
-    input:
-      context: "某 skill 连续 3 次漏判问题"
-    complexity: 3
-    expectedBehaviors:
-      - "收集该 skill 相关的所有反馈数据"
-      - "分析问题模式"
-      - "生成修改建议"
-      - "执行竞技场对比"
-      - "保持向后兼容"
-    expectedOutput:
-      - "升级后的 skill 文件"
-      - "变更说明"
-      - "A/B 对比报告"
-    successCriteria:
-      - "问题分析深入"
-      - "竞技场对比公正"
-      - "向后兼容"
-    qualityMetrics:
-      - "问题分析准确率 >= 90%"
-      - "兼容性保持率 = 100%"
-    maxDuration: 600
-
-  - name: "optimize-description"
-    input:
-      context: "某 skill 触发准确率仅 55%"
-    complexity: 1
-    expectedBehaviors:
-      - "分析误触发案例"
-      - "分析漏触发案例"
-      - "生成优化后的 description"
-      - "验证 CSO 格式"
-    expectedOutput:
-      - "优化后的 description"
-      - "A/B 测试报告"
-    successCriteria:
-      - "description 符合 CSO 格式"
-      - "触发准确率提升 >= 15%"
-    qualityMetrics:
-      - "CSO 格式符合率 = 100%"
-      - "准确率提升 >= 15%"
-    maxDuration: 300
-
-  - name: "frozen-zone-skip"
-    input:
-      context: "建议修改 constitution.md"
-    complexity: 1
-    expectedBehaviors:
-      - "识别目标文件属于冻结区"
-      - "停止自动执行"
-      - "仅记录建议"
-    expectedOutput:
-      - "建议记录"
-      - "冻结区提示"
-    successCriteria:
-      - "冻结区识别正确"
-      - "不执行修改"
-    qualityMetrics:
-      - "冻结区识别率 = 100%"
-      - "无违规修改率 = 100%"
-    maxDuration: 60
-
-  - name: "insufficient-evidence"
-    input:
-      context: "practice-log 中某模式仅出现 1 次"
-    complexity: 1
-    expectedBehaviors:
-      - "检测证据数量 < 3 条"
-      - "不触发锻造流程"
-      - "提示证据不足"
-    expectedOutput:
-      - "证据不足提示"
-    successCriteria:
-      - "证据数量检测正确"
-      - "不触发锻造"
-    qualityMetrics:
-      - "证据检测率 = 100%"
-    maxDuration: 60
-
-successCriteria:
-  passRate: 85
-  avgFieldCompletion: 90

package/plugins/benchmarks/templates/skills/startup-guard-template.yaml

@@ -1,103 +0,0 @@
-# AutoSpec Skill Benchmark Template - Startup-Guard
-# 适用于: 测试 startup-guard skill
-# init 后复制到 .autospec/benchmarks/ 后按需修改
-
-version: "1.0"
-name: "skill-startup-guard"
-description: "Startup-Guard Skill 基准测试 - 启动门禁"
-
-type: skill
-target: startup-guard
-
-testCases:
-  - name: "new-feature-guard"
-    input:
-      context: "实现 {feature-name} 功能"
-    complexity: 1
-    expectedBehaviors:
-      - "检测用户未使用 /autospec: 命令"
-      - "确认 .autospec/knowledge/ 存在"
-      - "情境识别"
-      - "推荐 /autospec:run 或 /autospec:explore"
-    expectedOutput:
-      - "门禁提示信息"
-      - "推荐命令列表"
-    successCriteria:
-      - "正确识别非命令触发"
-      - "情境识别准确"
-      - "推荐命令合理"
-    qualityMetrics:
-      - "门禁触发率 = 100%"
-      - "情境识别准确率 >= 90%"
-    maxDuration: 60
-
-  - name: "hotfix-guard"
-    input:
-      context: "生产环境发现严重 Bug，需要立即修复"
-    complexity: 1
-    expectedBehaviors:
-      - "识别紧急情境"
-      - "推荐 /autospec:run --workflow=hotfix"
-      - "说明热修复模式特点"
-    expectedOutput:
-      - "紧急响应提示"
-      - "热修复模式推荐"
-    successCriteria:
-      - "紧急情境识别正确"
-      - "热修复推荐合理"
-    qualityMetrics:
-      - "紧急识别率 = 100%"
-      - "热修复推荐准确率 = 100%"
-    maxDuration: 60
-
-  - name: "experiment-guard"
-    input:
-      context: "想验证一下新的技术方案是否可行"
-    complexity: 1
-    expectedBehaviors:
-      - "识别技术探索情境"
-      - "推荐 /autospec:run --workflow=experiment"
-    expectedOutput:
-      - "技术探索提示"
-      - "实验模式推荐"
-    successCriteria:
-      - "探索情境识别正确"
-    qualityMetrics:
-      - "探索识别率 = 100%"
-    maxDuration: 60
-
-  - name: "already-in-flow"
-    input:
-      context: "/autospec:run --workflow=waterfall 实现 {feature}"
-    complexity: 1
-    expectedBehaviors:
-      - "检测到用户已使用 /autospec: 命令"
-      - "本 skill 不生效"
-    successCriteria:
-      - "正确识别已走流程"
-      - "不重复提示门禁"
-    qualityMetrics:
-      - "命令识别率 = 100%"
-      - "无重复提示率 = 100%"
-    maxDuration: 30
-
-  - name: "knowledge-missing"
-    input:
-      context: "实现一个功能"
-      knowledgeMissing: true
-    complexity: 1
-    expectedBehaviors:
-      - "检测 .autospec/knowledge/ 不存在"
-      - "提示用户运行 autospec init"
-    expectedOutput:
-      - "初始化提示"
-      - "autospec init 命令"
-    successCriteria:
-      - "正确识别知识目录缺失"
-    qualityMetrics:
-      - "缺失识别率 = 100%"
-    maxDuration: 30
-
-successCriteria:
-  passRate: 95
-  avgFieldCompletion: 95