Repository navigation
v3.1.7
中文
新特性
- 新增
~ask命令:交互式需求澄清入口,一问一答厘清目标、方向、范围与约束,纯对话不创建文件。合并原~idea(方向探索)和~office(价值评估/范围收缩),提供统一入口 ~plan需求澄清阶段默认采用交互澄清模式(一问一答、选择题优先),已有代码库时先静默读代码再以选择题逐条确认,原"假设模式"改为快速通道(用户说"直接做,不用问我"时启用)~prd项目定位和维度探索阶段改为一问一答交互澄清,维度探索不再"AI 先给推荐让用户审查",改为逐条提问带推荐选项~build无方案包且任务 T2+ 时自动重定向到~plan,避免需求不清时直接写代码tasks.md模板完成标准改为"可独立验证的布尔条件"——任何人读完就能判断任务是否完成qa-review新增任务完成标准逐条核对步骤,验证不依赖主会话上下文
问题修复
- 修复
~idea兼容别名在脚本路由层失效问题(COMMAND_ALIASES缺少idea: 'ask'映射) - 统一所有命令 SKILL.md 中 AI 指令语气为第二人称"你"(原
~ask和~prd中混用"AI"第三人称) - 修复 guard 测试中断言与当前 guard.mjs 行为不一致的预存问题(高危命令由 deny 改为 warn 后测试未同步)
- 统一"可独立验证"语序(原"独立可验证"共 2 处不一致)
bootstrap.md/bootstrap-lite.md中 "SKILL" 加反引号,措辞"温和偏差拖小"改为"惯性偏差拖累"bootstrap-lite.md补充命令执行指引,避免精简模式下代理不清楚后续流程README_CN.md首页长句优化为更自然的表达
文档
- 全面更新 README 中英文版本中的命令表、别名和描述
杂项
zeroSideEffect标记添加注释说明设计意图
变更记录
完整更新记录:v3.1.5 beta (891bab2) ... v3.1.7 beta
feat新增~ask交互式需求澄清命令,合并~idea+~office功能feat~plan/~prd/~build全面采用交互澄清模式feat任务完成标准改为"可独立验证的布尔条件"fixCOMMAND_ALIASES补充idea: 'ask'映射fix统一所有命令 SKILL.md 语气为第二人称"你"fix修复 guard 测试预存断言不一致问题fix统一措辞、修复语言质量问题release同步v3.1.7 beta版本号
English
New Features
- New
~askcommand: interactive requirements clarification via structured Q&A. Covers direction, value judgment, scope, and premise challenging — pure conversation, no files created. Merges former~idea(exploration) and~office(value/scope review) into a single entry point ~plannow defaults to interactive clarification mode (one question at a time, multiple-choice preferred). Existing codebases get silent code reading first, then choice-based confirmation. Former "assumption mode" becomes a fast-track option ("just do it, don't ask me")~prdproject positioning and dimension exploration now use interactive Q&A. Dimension exploration no longer starts with "AI recommends first, user reviews" — it now asks one question at a time with recommended options~buildnow redirects to~planwhen no plan package exists and task is T2+, preventing premature coding with unclear requirementstasks.mdtemplate completion criteria updated to "independently verifiable boolean conditions" — anyone can judge task completion by reading the criteria aloneqa-reviewnow includes per-task completion criteria verification, independent of main session context
Bug Fixes
- Fixed
~ideacompatibility alias broken at script routing layer (COMMAND_ALIASESmissingidea: 'ask'mapping) - Unified AI directive voice to second-person "you" across all command SKILL.md files (previously mixed "AI" third-person in
~askand~prd) - Fixed stale guard test assertions mismatched with current guard.mjs behavior (HIGH_RISK commands changed from deny to warn)
- Unified word order for "independently verifiable" across all files (2 instances were inconsistent)
- Added backticks to "SKILL" in
bootstrap.md/bootstrap-lite.md; improved phrasing - Added command execution guidance to
bootstrap-lite.mdfor clarity in lightweight mode - Improved long-sentence readability in
README_CN.md
Documentation
- Comprehensively updated README (EN/CN) command table, aliases, and descriptions
Chores
- Added explanatory comment to
zeroSideEffectflag documenting design intent
Changelog
Full Changelog: v3.1.5 beta (891bab2) ... v3.1.7 beta
featadd~askinteractive clarification command, merge~idea+~officefeatadopt interactive clarification mode in~plan/~prd/~buildfeatupdate task completion criteria to independently verifiable boolean conditionsfixaddidea: 'ask'toCOMMAND_ALIASESfixunify all command SKILL.md voice to second-personfixsync stale guard test assertionsfixunify wording and fix language quality issuesreleasebump tov3.1.7 beta