通过测试与验证为先的方式,按序执行并实现已批准的规范提案。用于实施变更、应用提案、执行规范任务或按已批准计划构建。触发词包括 "speckit-implement", "speckit开发", "开发", "实施" "实现提案", "应用变更", "执行规范", "按顺序完成任务", "构建功能", "开始实施"。
通过测试与验证为先的方式,按序执行并实现已批准的规范提案。当用户使用"开发"、"实施"、"构建功能"等触发词时,自动检查任务清单、验证项目设置,并按计划执行实现任务。
/plugin marketplace add forztf/open-skilled-sdd/plugin install open-skilled-sdd@open-skilled-sdd-marketplaceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
assets/dockerignore.templateassets/eslintignore.templateassets/gitignore.templatereferences/context-loading-order.mdreferences/ignore-patterns.mdreferences/implementation-workflow.mdscripts/analyze-checklists.ps1scripts/check-prerequisites.ps1scripts/detect-project-setup.ps1$ARGUMENTS
您必须在继续之前考虑用户输入(如果不为空)。
scripts: sh: .specify/scripts/bash/check-prerequisites.sh --json ps: .specify/scripts/powershell/check-prerequisites.ps1 -Json
从仓库根目录运行 {SCRIPT} 并解析 FEATURE_DIR 和 AVAILABLE_DOCS 列表。所有路径必须是绝对的。对于参数中的单引号,如 "I'm Groot",使用转义语法:例如 'I'''m Groot'(或者如果可能的话使用双引号:"I'm Groot")。
检查检查表状态(如果 FEATURE_DIR/checklists/ 存在):
扫描 checklists/ 目录中的所有检查表文件
对于每个检查表,计数:
- [ ] 或 - [X] 或 - [x] 的行- [X] 或 - [x] 的行- [ ] 的行创建状态表:
| 检查表 | 总计 | 已完成 | 未完成 | 状态 |
|-----------|-------|-----------|------------|--------|
| ux.md | 12 | 12 | 0 | ✓ 通过 |
| test.md | 8 | 5 | 3 | ✗ 失败 |
| security.md | 6 | 6 | 0 | ✓ 通过 |
计算总体状态:
如果有任何检查表未完成:
如果所有检查表都完成:
加载和分析实现上下文:
项目设置验证:
检测和创建逻辑:
检查以下命令是否成功以确定仓库是否为 git 仓库(如果是则创建/验证 .gitignore):
git rev-parse --git-dir 2>/dev/null
检查是否存在 Dockerfile* 或 Docker 在 plan.md 中 → 创建/验证 .dockerignore
检查是否存在 .eslintrc* 或 eslint.config.* → 创建/验证 .eslintignore
检查是否存在 .prettierrc* → 创建/验证 .prettierignore
检查是否存在 .npmrc 或 package.json → 创建/验证 .npmignore(如果发布)
检查是否存在 terraform 文件 (*.tf) → 创建/验证 .terraformignore
检查是否需要 .helmignore(存在 helm 图表)→ 创建/验证 .helmignore
如果忽略文件已存在:验证它包含基本模式,仅追加缺失的关键模式 如果忽略文件缺失:创建包含检测技术的完整模式集
按技术的常见模式(来自 plan.md 技术栈):
node_modules/, dist/, build/, *.log, .env*__pycache__/, *.pyc, .venv/, venv/, dist/, *.egg-info/target/, *.class, *.jar, .gradle/, build/bin/, obj/, *.user, *.suo, packages/*.exe, *.test, vendor/, *.out.bundle/, log/, tmp/, *.gem, vendor/bundle/vendor/, *.log, *.cache, *.envtarget/, debug/, release/, *.rs.bk, *.rlib, *.prof*, .idea/, *.log, .env*build/, out/, .gradle/, .idea/, *.class, *.jar, *.iml, *.log, .env*build/, bin/, obj/, out/, *.o, *.so, *.a, *.exe, *.dll, .idea/, *.log, .env*build/, bin/, obj/, out/, *.o, *.a, *.so, *.exe, Makefile, config.log, .idea/, *.log, .env*.build/, DerivedData/, *.swiftpm/, Packages/.Rproj.user/, .Rhistory, .RData, .Ruserdata, *.Rproj, packrat/, renv/.DS_Store, Thumbs.db, *.tmp, *.swp, .vscode/, .idea/工具特定模式:
node_modules/, .git/, Dockerfile*, .dockerignore, *.log*, .env*, coverage/node_modules/, dist/, build/, coverage/, *.min.jsnode_modules/, dist/, build/, coverage/, package-lock.json, yarn.lock, pnpm-lock.yaml.terraform/, *.tfstate*, *.tfvars, .terraform.lock.hcl*.secret.yaml, secrets/, .kube/, kubeconfig*, *.key, *.crt解析 tasks.md 结构并提取:
按照任务计划执行实现:
实现执行规则:
进度跟踪和错误处理:
完成验证:
注意:此命令假设 tasks.md 中存在完整的任务分解。如果任务不完整或缺失,建议首先运行 speckit-tasks 以重新生成任务列表。
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.