@givolt/i3-composer-agent 2.2.3 → 2.3.1

package/.agents/plugins/marketplace.json

@@ -6,7 +6,7 @@
   "plugins": [
     {
       "name": "i3-composer-agent",
-      "version": "2.2.3",
+      "version": "2.3.1",
       "description": "覆盖需求、计划、开发、测试、审查与发布的软件工程技能，默认使用简体中文。",
       "source": {
         "source": "local",

package/.claude/commands/i3-prototype.md

@@ -0,0 +1,23 @@
+---
+description: "选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。"
+# i3-description-source: "description: Create, review and iterate interactive UI prototypes with template selection and approved baselines"
+---
+
+<!-- i3:zh-CN:start -->
+## 中文交互与输出
+
+默认使用简体中文回复，包括追问、选项说明、阶段进度、分析、规格、计划、验收、代码审查与发布报告。只有用户明确要求其他语言时才切换；输入中出现英文术语、代码或引用不构成切换语言的要求。
+下文的工程步骤、审批要求、退出条件和工具约束完整有效。英文示例中的人类可读标题、说明和占位内容在实际输出时应译为中文；保留原有信息与验证标准，不删减方法。
+Skill/Command/Agent 标识、文件路径、命令参数、代码/API 名称、工具调用参数、JSON 键/类型/枚举、机器状态、测试期望字面量及引用原文保持原样。生成文档正文用中文，SPEC.md、tasks/plan.md、tasks/todo.md、CONSTRAINTS.md 等约定路径不改名。
+严重程度按原规则判断，面向用户可写“严重（Critical）”“必须修复（Required）”“小问题（Nit）”“可选（Optional）”“建议考虑（Consider）”“仅供参考（FYI）”；High/Medium/Low、APPROVE/REQUEST CHANGES 和 GO/NO-GO 保留其原始语义与机器值。不得因翻译降低门槛，角色不能互相委派。
+
+适用说明：选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。
+<!-- i3:zh-CN:end -->
+
+Invoke the i3-ui-prototype skill.
+
+$ARGUMENTS
+
+With no arguments, use create. Supported operations: create, revise, inspect, confirm-baseline.
+Default to professional enhancement mode. Detect UI/UX Pro Max in the current host; if unavailable, offer installation or explicit template mode. Never silently downgrade.
+Use the skill workflow for template selection, business mapping, professional advice, real browser verification and version-specific human confirmation. Preserve approved snapshots during iteration.

package/.claude/commands/i3-review.md

@@ -24,5 +24,5 @@ Review the current changes (staged or recent commits) across all five axes:
 4. **Security** — Input validated? Secrets safe? Auth checked? (Use i3-security-and-hardening skill)
 5. **Performance** — No N+1 queries? No unbounded ops? (Use i3-performance-optimization skill)
 
-Categorize findings as Critical, Important, or Suggestion.
+Categorize findings as Critical, Required, Optional, or Nit.
 Output a structured review with specific file:line references and fix recommendations.

package/.claude/commands/i3-ship.md

@@ -43,7 +43,7 @@ Constraints (from Claude Code's subagent model):
 
 Once all three reports are back, the main agent (not a sub-persona) synthesizes them:
 
-1. **Code Quality** — Aggregate Critical/Important findings from `code-reviewer` and any failing tests, lint, or build output. Resolve duplicates between reviewers.
+1. **Code Quality** — Aggregate Critical/Required findings from `code-reviewer` and any failing tests, lint, or build output. Resolve duplicates between reviewers.
 2. **Security** — Promote any Critical/High `security-auditor` findings to launch blockers. Cross-reference with `code-reviewer`'s security axis.
 3. **Performance** — Pull from `code-reviewer`'s performance axis; cross-check Core Web Vitals if applicable.
 4. **Accessibility** — Verify keyboard nav, screen reader support, contrast (not covered by the three personas — handle directly here, or invoke the accessibility checklist).

package/.claude-plugin/marketplace.json

@@ -8,7 +8,7 @@
   "plugins": [
     {
       "name": "i3-composer-agent",
-      "version": "2.2.3",
+      "version": "2.3.1",
       "source": "./",
       "description": "覆盖需求、计划、开发、测试、审查与发布的软件工程技能，默认使用简体中文。",
       "homepage": "https://github.com/49780522/i3-composer-agent",

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "i3-composer-agent",
-  "version": "2.2.3",
+  "version": "2.3.1",
   "description": "覆盖需求、计划、开发、测试、审查与发布的软件工程技能，默认使用简体中文。",
   "author": {
     "name": "Givolt"

package/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "i3-composer-agent",
-  "version": "2.2.3",
+  "version": "2.3.1",
   "description": "覆盖需求、计划、开发、测试、审查与发布的软件工程技能，默认使用简体中文。",
   "author": {
     "name": "Givolt"
@@ -12,7 +12,7 @@
   "interface": {
     "displayName": "i3 Composer Agent",
     "shortDescription": "中文软件工程全流程技能",
-    "longDescription": "i3 Composer Agent 提供 25 项工程技能，覆盖需求、计划、实现、测试、审查与发布，默认使用简体中文。",
+    "longDescription": "i3 Composer Agent 提供 26 项工程技能，覆盖需求、计划、实现、测试、审查与发布，默认使用简体中文。",
     "developerName": "Givolt",
     "category": "Productivity",
     "capabilities": [

package/.gemini/commands/i3-prototype.toml

@@ -0,0 +1,23 @@
+description = "选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。"
+# i3-description-source: "description = \"Create, review and iterate interactive UI prototypes with template selection and approved baselines\""
+
+prompt = """
+<!-- i3:zh-CN:start -->
+## 中文交互与输出
+
+默认使用简体中文回复，包括追问、选项说明、阶段进度、分析、规格、计划、验收、代码审查与发布报告。只有用户明确要求其他语言时才切换；输入中出现英文术语、代码或引用不构成切换语言的要求。
+下文的工程步骤、审批要求、退出条件和工具约束完整有效。英文示例中的人类可读标题、说明和占位内容在实际输出时应译为中文；保留原有信息与验证标准，不删减方法。
+Skill/Command/Agent 标识、文件路径、命令参数、代码/API 名称、工具调用参数、JSON 键/类型/枚举、机器状态、测试期望字面量及引用原文保持原样。生成文档正文用中文，SPEC.md、tasks/plan.md、tasks/todo.md、CONSTRAINTS.md 等约定路径不改名。
+严重程度按原规则判断，面向用户可写“严重（Critical）”“必须修复（Required）”“小问题（Nit）”“可选（Optional）”“建议考虑（Consider）”“仅供参考（FYI）”；High/Medium/Low、APPROVE/REQUEST CHANGES 和 GO/NO-GO 保留其原始语义与机器值。不得因翻译降低门槛，角色不能互相委派。
+
+适用说明：选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。
+<!-- i3:zh-CN:end -->
+
+Invoke the i3-ui-prototype skill.
+
+$ARGUMENTS
+
+With no arguments, use create. Supported operations: create, revise, inspect, confirm-baseline.
+Default to professional enhancement mode. Detect UI/UX Pro Max in the current host; if unavailable, offer installation or explicit template mode. Never silently downgrade.
+Use the skill workflow for template selection, business mapping, professional advice, real browser verification and version-specific human confirmation. Preserve approved snapshots during iteration.
+"""

package/.gemini/commands/i3-review.toml

@@ -23,6 +23,6 @@ Review the current changes (staged or recent commits) across all five axes:
 4. **Security** — Input validated? Secrets safe? Auth checked? (Use i3-security-and-hardening skill)
 5. **Performance** — No N+1 queries? No unbounded ops? (Use i3-performance-optimization skill)
 
-Categorize findings as Critical, Important, or Suggestion.
+Categorize findings as Critical, Required, Optional, or Nit.
 Output a structured review with specific file:line references and fix recommendations.
 """

package/.gemini/commands/i3-ship.toml

@@ -40,7 +40,7 @@ Constraints (from Gemini CLI's subagent model):
 
 Once all three reports are back, the main agent (not a sub-persona) synthesizes them:
 
-1. **Code Quality** — Aggregate Critical/Important findings from `code-reviewer` and any failing tests, lint, or build output. Resolve duplicates between reviewers.
+1. **Code Quality** — Aggregate Critical/Required findings from `code-reviewer` and any failing tests, lint, or build output. Resolve duplicates between reviewers.
 2. **Security** — Promote any Critical/High `security-auditor` findings to launch blockers. Cross-reference with `code-reviewer`'s security axis.
 3. **Performance** — Pull from `code-reviewer`'s performance axis; cross-check Core Web Vitals if applicable.
 4. **Accessibility** — Verify keyboard nav, screen reader support, contrast (not covered by the three personas — handle directly here, or invoke the accessibility checklist).

package/README.md

@@ -1,4 +1,4 @@
-# i3 Composer Agent v2.2.3
+# i3 Composer Agent v2.3.1
 
 **Intelligence. Integration. Innovation.**
 
@@ -25,8 +25,8 @@ WorkBuddy 在第 2 项。上下键移动，空格勾选/取消，回车确认；
 搜索、翻页和命令参数预选均保留选择。搜索外 `q`、Esc 或 Ctrl-C 取消。
 确认工具后选择项目/全局范围并确认安装。
 
-全部目标均安装 25 个技能、共享参考资料，以及完整评测/Hook/脚本配套。
-全部目标同时提供 9 个命令和 4 个 Agent 角色提示。
+全部目标均安装 26 个技能、共享参考资料，以及完整评测/Hook/脚本配套。
+全部目标同时提供 10 个命令和 4 个 Agent 角色提示。
 CodeBuddy、WorkBuddy、Claude Code 使用原生命令入口；其他工具通过 Skills 机制使用，
 命令和角色文件可作为资料读取，是否自动发现取决于宿主。
 目录预设来自固定的 [Skills CLI 工具登记](https://github.com/vercel-labs/skills/blob/48dc9e8eb8aa19040e91cb90ba0b642c9cbad57e/src/agents.ts)
@@ -161,6 +161,7 @@ CodeBuddy、WorkBuddy、Claude Code 原生安装使用 `/i3-spec`、`/i3-plan`
 | /i3-composer-agent:i3-code-simplify | 简化代码，降低复杂度，提高可读性与维护性，并保持既有行为不变。 |
 | /i3-composer-agent:i3-constraints | 访谈并定义项目质量门槛，提供合理默认值，写入 CONSTRAINTS.md 并落实约束。 |
 | /i3-composer-agent:i3-plan | 把工作拆成带验收标准、验证步骤和依赖顺序的小任务。 |
+| /i3-composer-agent:i3-prototype | 选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。 |
 | /i3-composer-agent:i3-review | 从正确性、可读性、架构、安全和性能五个维度审查代码。 |
 | /i3-composer-agent:i3-ship | 并行调用三个专业角色执行发布前检查，主会话汇总 GO/NO-GO 决策与回滚计划。 |
 | /i3-composer-agent:i3-spec | 先编写结构化规格、目标、范围与验收标准，再进入实现。 |
@@ -169,8 +170,16 @@ CodeBuddy、WorkBuddy、Claude Code 原生安装使用 `/i3-spec`、`/i3-plan`
 
 `/i3-build auto` 按已获批准的任务计划执行完整流程。
 插件调用为 `/i3-composer-agent:i3-build auto`，原审批边界和验证条件仍有效。
+`/i3-review` 在主会话中使用代码审查技能；`/i3-test` 在主会话中使用测试驱动开发技能，
+这两个命令不会自动启动同名专业角色。测试规范的编写与 TDD 执行应按各自任务处理。
 `/i3-ship` 在宿主支持时并行运行三个专业角色；不可用时读取角色提示依次审查，
-由主会话汇总 GO/NO-GO 决策，回滚计划必需，严重问题默认阻止发布。
+由主会话汇总 GO/NO-GO 决策，回滚计划必需。代码审查使用 Critical、Required、Optional、Nit
+四级；Critical 和 Required 问题默认阻止发布，各专业角色仍保留自己的报告字段和判断规则。
+
+`/i3-prototype` 调用 `i3-ui-prototype` 技能，支持先选模板，再按业务需求生成原型。默认采用专业增强模式，
+UI/UX Pro Max 是需要单独安装的外部设计能力，不计入本包的技能数量；未安装时提示安装，
+用户也可明确选择模板模式继续。企业后台模板可用于人工测试，AI 工作台、数字孪生大屏和
+移动端模板仍在规划中。原型验收、后续迭代约束与项目需求需要单独确认；本包尚未实现企业 SOP 自动编排。
 
 ## Skills
 
@@ -199,7 +208,8 @@ CodeBuddy、WorkBuddy、Claude Code 原生安装使用 `/i3-spec`、`/i3-plan`
 | [i3-shipping-and-launch](skills/i3-shipping-and-launch/SKILL.md) | 检查生产发布条件与回滚策略。适用于上线准备、发布清单、监控、分阶段部署和回滚方案，确保风险有依据且可恢复。 |
 | [i3-source-driven-development](skills/i3-source-driven-development/SKILL.md) | 以官方文档和权威来源支撑实现决策。适用于核对框架/库用法、验证方案、查证 API 和避免过时实现，并保留来源证据。 |
 | [i3-spec-driven-development](skills/i3-spec-driven-development/SKILL.md) | 在实现前形成结构化规格。适用于新项目、功能或重大改动、PRD、需求澄清、目标与范围、场景和验收标准，以及将需求拆为独立可测试能力。 |
-| [i3-test-driven-development](skills/i3-test-driven-development/SKILL.md) | 执行测试先行的红绿重构循环。适用于实现逻辑、修复缺陷、修改行为、复现报错及证明改动有效；缺陷修复使用先证明失败的 Prove-It 模式。 |
+| [i3-test-driven-development](skills/i3-test-driven-development/SKILL.md) | 执行测试先行的红绿重构循环。适用于实现逻辑、修复缺陷、修改行为、复现报错及证明改动有效；也用于为函数补测试、修复或修改直接粘贴的代码，即使请求未提到测试；缺陷修复使用先证明失败的 Prove-It 模式。 |
+| [i3-ui-prototype](skills/i3-ui-prototype/SKILL.md) | 根据业务需求选择 UI 模板，生成、评审并迭代交互原型。适用于选择原型模板、制作可操作审查原型、检查原型一致性及维护已确认的原型基线。 |
 | [i3-using-agent-skills](skills/i3-using-agent-skills/SKILL.md) | 发现并调用合适的工程技能。适用于开始会话、判断任务意图、选择技能或工作流；作为元技能协调其他技能的发现和调用。 |
 
 ## Agent 角色

package/agents/code-reviewer.md

@@ -110,5 +110,5 @@ Categorize every finding, using the same severity labels as the `i3-code-review-
 ## Composition
 
 - **Invoke directly when:** the user asks for a review of a specific change, file, or PR.
-- **Invoke via:** `/i3-review` (single-perspective review) or `/i3-ship` (parallel fan-out alongside `security-auditor` and `test-engineer`).
+- **Invoke via:** `/i3-ship` (parallel fan-out alongside `security-auditor` and `test-engineer`). `/i3-review` runs the `i3-code-review-and-quality` skill in the main session and does not spawn this persona.
 - **Do not invoke from another persona.** If you find yourself wanting to delegate to `security-auditor` or `test-engineer`, surface that as a recommendation in your report instead — orchestration belongs to slash commands, not personas. See [docs/agents.md](../docs/agents.md).

package/agents/test-engineer.md

@@ -103,5 +103,5 @@ When analyzing test coverage:
 ## Composition
 
 - **Invoke directly when:** the user asks for test design, coverage analysis, or a Prove-It test for a specific bug.
-- **Invoke via:** `/i3-test` (TDD workflow) or `/i3-ship` (parallel fan-out for coverage gap analysis alongside `code-reviewer` and `security-auditor`).
+- **Invoke via:** `/i3-ship` (parallel fan-out for coverage gap analysis alongside `code-reviewer` and `security-auditor`). `/i3-test` runs the `i3-test-driven-development` skill in the main session and does not spawn this persona.
 - **Do not invoke from another persona.** Recommendations to add tests belong in your report; the user or a slash command decides when to act on them. See [docs/agents.md](../docs/agents.md).

package/bin/native-layout.cjs

@@ -27,6 +27,8 @@ function nativeScript(text, relative) {
   if (/-test\.js$/.test(relative)) return text;
   if (relative === 'scripts/validate-skills.js') {
     text = replaceOnce(text, "const SKILLS_DIR = path.resolve(__dirname, '..', 'skills');", setup + "\nconst SKILLS_DIR = INSTALLATION ? INSTALLATION.skillsDirectory : path.resolve(__dirname, '..', 'skills');", relative);
+    text = replaceOnce(text, "const AGENTS_DIR = path.resolve(__dirname, '..', 'agents');", "const AGENTS_DIR = INSTALLATION ? path.join(INSTALLATION.root, 'agents') : path.resolve(__dirname, '..', 'agents');", relative);
+    text = replaceOnce(text, 'fs.readdirSync(AGENTS_DIR)', "(INSTALLATION ? INSTALLATION.catalog.agents.map(name => name + '.md') : fs.readdirSync(AGENTS_DIR))", relative);
     return replaceOnce(text, 'fs.readdirSync(SKILLS_DIR)', '(INSTALLATION ? INSTALLATION.catalog.skills : fs.readdirSync(SKILLS_DIR))', relative);
   }
   const rootDeclaration = relative === 'scripts/run-evals.js'

package/commands/i3-prototype.toml

@@ -0,0 +1,23 @@
+description = "选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。"
+# i3-description-source: "description = \"Create, review and iterate interactive UI prototypes with template selection and approved baselines\""
+
+prompt = """
+<!-- i3:zh-CN:start -->
+## 中文交互与输出
+
+默认使用简体中文回复，包括追问、选项说明、阶段进度、分析、规格、计划、验收、代码审查与发布报告。只有用户明确要求其他语言时才切换；输入中出现英文术语、代码或引用不构成切换语言的要求。
+下文的工程步骤、审批要求、退出条件和工具约束完整有效。英文示例中的人类可读标题、说明和占位内容在实际输出时应译为中文；保留原有信息与验证标准，不删减方法。
+Skill/Command/Agent 标识、文件路径、命令参数、代码/API 名称、工具调用参数、JSON 键/类型/枚举、机器状态、测试期望字面量及引用原文保持原样。生成文档正文用中文，SPEC.md、tasks/plan.md、tasks/todo.md、CONSTRAINTS.md 等约定路径不改名。
+严重程度按原规则判断，面向用户可写“严重（Critical）”“必须修复（Required）”“小问题（Nit）”“可选（Optional）”“建议考虑（Consider）”“仅供参考（FYI）”；High/Medium/Low、APPROVE/REQUEST CHANGES 和 GO/NO-GO 保留其原始语义与机器值。不得因翻译降低门槛，角色不能互相委派。
+
+适用说明：选择模板并生成、评审和迭代交互原型，维护已确认的 UI 基线。
+<!-- i3:zh-CN:end -->
+
+Invoke the i3-ui-prototype skill.
+
+$ARGUMENTS
+
+With no arguments, use create. Supported operations: create, revise, inspect, confirm-baseline.
+Default to professional enhancement mode. Detect UI/UX Pro Max in the current host; if unavailable, offer installation or explicit template mode. Never silently downgrade.
+Use the skill workflow for template selection, business mapping, professional advice, real browser verification and version-specific human confirmation. Preserve approved snapshots during iteration.
+"""

package/commands/i3-review.toml

@@ -23,6 +23,6 @@ Review the current changes (staged or recent commits) across all five axes:
 4. **Security** — Input validated? Secrets safe? Auth checked? (Use i3-security-and-hardening skill)
 5. **Performance** — No N+1 queries? No unbounded ops? (Use i3-performance-optimization skill)
 
-Categorize findings as Critical, Important, or Suggestion.
+Categorize findings as Critical, Required, Optional, or Nit.
 Output a structured review with specific file:line references and fix recommendations.
 """

package/commands/i3-ship.toml

@@ -40,7 +40,7 @@ Constraints (from CLI's subagent model):
 
 Once all three reports are back, the main agent (not a sub-persona) synthesizes them:
 
-1. **Code Quality** — Aggregate Critical/Important findings from `code-reviewer` and any failing tests, lint, or build output. Resolve duplicates between reviewers.
+1. **Code Quality** — Aggregate Critical/Required findings from `code-reviewer` and any failing tests, lint, or build output. Resolve duplicates between reviewers.
 2. **Security** — Promote any Critical/High `security-auditor` findings to launch blockers. Cross-reference with `code-reviewer`'s security axis.
 3. **Performance** — Pull from `code-reviewer`'s performance axis; cross-check Core Web Vitals if applicable.
 4. **Accessibility** — Verify keyboard nav, screen reader support, contrast (not covered by the three personas — handle directly here, or invoke the accessibility checklist).

package/docs/agents.md

@@ -42,10 +42,10 @@ Pick this when you want one perspective on the current change and the user is in
 ### Slash command (single persona behind it)
 Pick this when there's a repeatable workflow you'd otherwise re-explain every time.
 
-- `/i3-review` → wraps `code-reviewer` with the project's review skill
-- `/i3-test` → wraps `test-engineer` with TDD skill
 - `/i3-webperf` → wraps `web-performance-auditor` for performance-focused audits on web apps
 
+`/i3-review` and `/i3-test` are not persona wrappers. They run the `i3-code-review-and-quality` and `i3-test-driven-development` skills in the main session, where the conversation's context is. To get `code-reviewer` or `test-engineer` in a fresh context, invoke the persona directly or run `/i3-ship`.
+
 ### Slash command (orchestrator — fan-out)
 Pick this only when **independent** investigations can run in parallel and produce reports that a single agent then merges.
 

package/docs/codebuddy-setup.md

@@ -9,8 +9,8 @@ npx @givolt/i3-composer-agent --codebuddy
 ```
 
 向导预选 CodeBuddy，可增加其他工具，再选择项目或全局范围。
-项目安装提供 `.codebuddy/skills/`（25 个技能）、`references/`、
-`commands/`（9 个命令）和 `agents/`（4 个角色）。
+项目安装提供 `.codebuddy/skills/`（26 个技能）、`references/`、
+`commands/`（10 个命令）和 `agents/`（4 个角色）。
 同级 `i3-composer-agent/` 是配套工具根，集中提供 `package.json`、`evals/`、`hooks/`、`scripts/`、`docs/`、`references/` 和校验输入 `adapters/`，
 只有一份实际技能、命令和角色，评测与宿主共用。
 全局使用 `~/.codebuddy/` 的相同结构。默认中文说明和回复；技能/命令标识不改名。

package/docs/codex-setup.md

@@ -7,7 +7,7 @@ npx @givolt/i3-composer-agent --codex
 ```
 
 项目范围写入 `.agents/skills/`，全局写入 `~/.agents/skills/`，并安装同级 `i3-composer-agent/references/`。
-提供全部 25 个 `i3-` 技能，中文选择说明与默认中文回复在每个入口独立生效。
+提供全部 26 个 `i3-` 技能，中文选择说明与默认中文回复在每个入口独立生效。
 同级 `evals/`、`hooks/`、`scripts/`、`docs/` 各自以 `i3-composer-agent/` 分组资料。
 命令与角色文件也会提供，供按需阅读，其自动发现由宿主决定。
 从项目目录执行 `node .agents/i3-composer-agent/scripts/run-evals.js --min-rank1 95`

package/docs/copilot-setup.md

@@ -12,6 +12,8 @@ This guide covers Copilot in VS Code. For the standalone `copilot` command-line
 
 **What an install actually gives you:** the skills. Each installed skill becomes a slash command named after its frontmatter `name` — `/i3-spec-driven-development`, `/i3-test-driven-development`, and so on. `npx skills add /path/to/i3-composer-agent` and the manual copy below both install skills only. Neither of those two routes copies this repo's short lifecycle wrappers (`/i3-spec`, `/i3-plan`, `/i3-build`, `/i3-test`, `/i3-review`, `/i3-ship`) — those are Claude Code commands living in `.claude/commands/`. Use the full skill names, or add your own aliases — see [Lifecycle workflows](#lifecycle-workflows).
 
+> **`/i3-spec`, `/i3-plan`, `/i3-build`, `/i3-test`, `/i3-review`, `/i3-code-simplify`, `/i3-webperf` and `/i3-ship` never appear in Copilot.** They are Claude Code commands, and Copilot does not read `.claude/commands/`. Not seeing them after an install is expected, not a missing step. Invoke the skill by its own name instead — `/i3-spec-driven-development` for `/i3-spec`, `/i3-planning-and-task-breakdown` for `/i3-plan`, and so on; the full mapping is in [Lifecycle Workflows](#lifecycle-workflows). On VS Code you can add [short aliases](#optional-short-spec-style-aliases) yourself if you want the short forms.
+
 ## Setup
 
 ### Copilot Instructions
@@ -153,6 +155,8 @@ Write the body yourself, or take it from this table, rather than copying `.claud
 
 ### If the skill slash commands don't appear
 
+First rule out the expected case: if the only commands you miss are the short lifecycle wrappers (`/i3-spec`, `/i3-plan`, `/i3-build`, `/i3-test`, `/i3-review`, `/i3-ship`), nothing is wrong — those are Claude Code commands and never surface in Copilot (see the note at the top). The steps below are for when the skills' own names, such as `/i3-spec-driven-development`, are missing too.
+
 Work through these in order — `/i3-spec-driven-development` missing and `/i3-spec` missing have different causes:
 
 1. **Check where the skills landed.** Each one needs its own directory containing a `SKILL.md`. The project-level paths this guide uses are `.github/skills/`, `.claude/skills/` and `.agents/skills/` — but an installer run with a global/user flag writes outside the workspace instead (`~/.agents/skills/` and equivalents), so a skill that isn't in any of the three may simply be installed personally. Confirm the actual path against the [VS Code agent skills docs](https://code.visualstudio.com/docs/agent-customization/agent-skills), then re-install into the project if you wanted it project-scoped.

package/docs/cursor-setup.md

@@ -10,6 +10,8 @@
 
 How to wire [i3-composer-agent](../README.md) into **Cursor** using current, supported project context — not legacy monolith files or Kaizen-specific layouts.
 
+Every `.cursor/` path in this guide lives in **your** project (or under `~/.cursor/` for user scope). This repository does not ship a `.cursor/` directory: you create it and sync the skills into it, which is what [step 1](#1-install-skills-into-cursorskills) does.
+
 ---
 
 ## What Cursor supports today
@@ -213,6 +215,7 @@ Files under `i3-composer-agent/agents/` (e.g. code reviewer persona) are **not**
 | Stale workflow | Re-`rsync` from `i3-composer-agent/skills/` |
 | Duplicate instructions | Remove skill content from rules; keep one source |
 | Wrong skill picked | Narrow `description` in custom skills; nudge in chat |
+| "An unexpected error occurred on our servers" right after attaching a skill with `/`, and every later message in that chat fails too | Known Cursor-side issue, tracked in [upstream agent-skills #494](https://github.com/addyosmani/agent-skills/issues/494): Cursor pulls the attached `SKILL.md` into the request and the failure happens on its side, not in the skill. Start a new chat and reference the skill by name in plain text ("Use the i3-test-driven-development skill to …") instead of attaching it. |
 
 ---
 

package/docs/other-hosts.md

@@ -1,15 +1,50 @@
-# Other hosts
+# 其他宿主
 
-<!-- i3:zh-guide:start -->
-## 中文使用说明
+这些宿主尚未完成当前版本的真实宿主与模型执行验证。安装文件、宿主发现技能和模型正确执行是三项独立检查；请在测试项目中验证后再用于正式工作。
 
-按下文完整工程步骤、检查项与验证条件执行。根据任务选择适用步骤，保留证据、风险和退出条件；正文示例中的英文标签在面向用户输出时译为中文。
+技能包包含 26 个技能及共享资源。接入时应保留完整的技能、引用与配套工具，不能只复制 `SKILL.md`。
 
-默认使用简体中文解释、追问及生成报告；用户明确要求其他语言时可切换。原方法正文、引用、阈值和审批条件完整保留。代码标识、路径、命令、工具参数、JSON 键/枚举与原文引用保持原样。
-<!-- i3:zh-guide:end -->
+## 使用本地分发目录
 
-These hosts install the pack but are not in the README's install section, because no maintainer has run the current release on them. Each entry is one line: the host and its install command. To add one, see [Adding a Host Guide](../CONTRIBUTING.md#adding-a-host-guide).
+以下命令使用完整分发包解压目录里的 CLI，适用于本地测试和离线接入。`/absolute/path/to/package` 均替换为实际解压目录；已发布版本也可使用 `npx @givolt/i3-composer-agent@2.3.1` 的对应入口。
 
-Skills are plain `SKILL.md` folders, so any agent that reads the i3 Composer Agent layout can load them. See [getting-started.md](getting-started.md).
+### fx
 
-- **Oh My Pi**: `omp plugin marketplace add /path/to/i3-composer-agent`, then `omp plugin install i3-composer-agent@i3-composer-agent`
+在业务项目目录执行：
+
+```bash
+node /absolute/path/to/package/bin/i3-composer-agent.cjs --agents fx --scope project --yes
+```
+
+原生目标使用 `.fx/skills/`；全局安装可将 `--scope project` 改成 `--scope global`。上游记录要求 fx 0.0.5 或更新版本；当前版本的宿主发现与调用仍需人工验证。
+
+### Pi
+
+在业务项目目录执行：
+
+```bash
+node /absolute/path/to/package/bin/i3-composer-agent.cjs --agents pi --scope project --yes
+```
+
+原生目标使用 `.agents/skills/`；全局安装可将 `--scope project` 改成 `--scope global`。当前版本的宿主发现与调用仍需人工验证。
+
+### Dojo Workspace
+
+上游接入方式使用标准 `.agents/skills/` 布局。i3 CLI 目前没有专用 Dojo 目标；应先核对实际宿主的本地技能接入协议、共享资源路径和技能发现行为，再确定安装方式。不要将其他宿主目标的安装成功视为 Dojo 已验证。
+
+### Oh My Pi
+
+先将本地分发包导出为完整插件目录：
+
+```bash
+node /absolute/path/to/package/bin/i3-composer-agent.cjs --extract --dir ./i3-composer-agent
+```
+
+然后使用实际导出目录的绝对路径注册：
+
+```text
+omp plugin marketplace add /absolute/path/to/i3-composer-agent
+omp plugin install i3-composer-agent@i3-composer-agent
+```
+
+这是待人工验证的插件接入路径，i3 CLI 目前没有专用 Oh My Pi 原生目标。注册后需另行检查技能、命令和角色是否被宿主发现，以及模型能否正确调用；Hook 需要按宿主协议单独启用。

package/docs/workbuddy-setup.md

@@ -10,8 +10,8 @@ npx @givolt/i3-composer-agent --agents codebuddy,workbuddy --scope global --yes
 
 向导中 WorkBuddy 位于第 2 项，预选后仍可多选。全局目录：
 
-- `~/.workbuddy/skills/`：25 个 `i3-` 技能，选择说明和默认回复中文。
-- `~/.workbuddy/commands/`：9 个 `/i3-*` 短命令。
+- `~/.workbuddy/skills/`：26 个 `i3-` 技能，选择说明和默认回复中文。
+- `~/.workbuddy/commands/`：10 个 `/i3-*` 短命令。
 - `~/.workbuddy/agents/`：4 个专业角色。
 - `~/.workbuddy/i3-composer-agent/references/`：共享参考。
 - `~/.workbuddy/i3-composer-agent/docs/agents.md`：角色说明。

package/evals/README.md

@@ -73,6 +73,8 @@ node .codebuddy/i3-composer-agent/scripts/run-evals.js --behavioral i3-test-driv
 可能编辑文件、运行命令、访问网络；各样例还可能需要自身的语言运行环境。
 评分依据执行轨迹或对话与 `expectations`，不会把“声称完成”当成实际执行证据。
 模型输出按不可信数据评分，执行和评分均有限时。临时项目完成后清理。
+每项评分必须提供非空白的 `evidence`；缺失或空白证据会使该次评分无效，原始输出保留为
+`.grading.raw.txt`，不能视为行为验收通过。
 
 `--locale en` 运行英文用例，省略该参数运行两种语言。可通过 `I3_EVAL_MODEL`
 和 `I3_EVAL_GRADER_MODEL` 指定执行/评分模型；报告同时记录宿主和模型信息。

package/evals/cases/i3-documentation-and-adrs.json

@@ -39,6 +39,24 @@
         "Trade-offs and rejected options are recorded, not just the winning choice",
         "The ADR records a status from the skill's lifecycle (Proposed, Accepted, Superseded or Deprecated) and a date"
       ]
+    },
+    {
+      "id": 4,
+      "kind": "execution",
+      "prompt": "Review README.md for operators onboarding to this batch queue. Make the prose clearer, check it against queue.js, and keep the operational details and caveats accurate. Do not change the implementation.",
+      "expected_output": "A concise operational README grounded in the inspected implementation, with unsupported assurances removed or flagged, limits and caveats preserved, and the unverified command identified",
+      "files": [
+        "i3-documentation-and-adrs/prose-review"
+      ],
+      "expectations": [
+        "The final README identifies queuectl submit jobs.json --max-batch-size 50 as an unverified documented example because the fixture contains neither the CLI nor jobs.json; it does not invent, silently certify, or replace that interface",
+        "The final README preserves the 50-job batch limit, 72-hour retention period, and at most three retries, distinguishing those units correctly",
+        "The final README preserves the operational caveat that failed jobs require operator review and queue state is not durable across a process restart",
+        "The unsupported claims of guaranteed zero job loss and no need for monitoring are removed or explicitly marked as unsubstantiated rather than restated as facts",
+        "The promotional opening is replaced with a concrete description of the queue without invented users, benchmarks, certifications, or capabilities",
+        "The execution trace shows the agent inspecting queue.js and comparing the README with its batch, retention, retry, and in-memory-state implementation before editing",
+        "queue.js remains unchanged, and the agent does not install a tool or send the fixture to a hosted service"
+      ]
     }
   ]
 }

package/evals/cases/i3-test-driven-development.json

@@ -13,6 +13,14 @@
       {
         "prompt": "What tests should cover this new parsing logic before I write it?",
         "top_k": 3
+      },
+      {
+        "prompt": "Add unit tests for this function. There are none yet.",
+        "top_k": 3
+      },
+      {
+        "prompt": "Fix this bug: splitCents(100, 3) returns [33, 33, 33] and a cent goes missing. Here is the function.",
+        "top_k": 3
       }
     ],
     "negative": [

package/evals/cases/i3-ui-prototype.json

@@ -0,0 +1,49 @@
+{
+  "skill_name": "i3-ui-prototype",
+  "trigger": {
+    "positive": [
+      {
+        "prompt": "Choose a UI prototype template and generate an interactive approval prototype for review",
+        "top_k": 1
+      },
+      {
+        "prompt": "Generate a reviewable interactive UI prototype from business requirements",
+        "top_k": 1
+      },
+      {
+        "prompt": "Inspect prototype fidelity against its approved prototype baseline",
+        "top_k": 1
+      },
+      {
+        "prompt": "Iterate the interactive prototype using review feedback and preserve the approved baseline",
+        "top_k": 1
+      }
+    ],
+    "negative": [
+      {
+        "prompt": "Build an accessible modal component with keyboard support",
+        "owner": "i3-frontend-ui-engineering"
+      },
+      {
+        "prompt": "Optimize slow database indexes",
+        "owner": "i3-performance-optimization"
+      }
+    ]
+  },
+  "evals": [
+    {
+      "id": 1,
+      "prompt": "Create an interactive application approval prototype from the supplied business brief. Use template mode explicitly, compare the templates and generate a review candidate. Do not approve it yourself.",
+      "expected_output": "A runnable business-specific prototype with traceability and pending-review status",
+      "files": [
+        "i3-ui-prototype"
+      ],
+      "expectations": [
+        "Selected template and template mode are recorded without claiming UI/UX Pro Max was used",
+        "Business fields, role visibility, status transitions and review scenarios follow the supplied brief",
+        "A runnable prototype and verification evidence are produced; absent browser evidence is marked pending",
+        "The prototype remains pending human review and no approved baseline is invented"
+      ]
+    }
+  ]
+}

package/evals/fixtures/i3-documentation-and-adrs/prose-review/README.md

@@ -0,0 +1,19 @@
+# Batch Queue
+
+Batch Queue is an innovative, enterprise-grade solution that seamlessly unlocks
+the power of batch processing. In today's fast-paced landscape, it is important
+to note that this robust queue empowers operators to take work to the next level.
+
+## Submit a batch
+
+```sh
+queuectl submit jobs.json --max-batch-size 50
+```
+
+A batch contains at most 50 jobs. Queue entries expire after 72 hours. Failed
+jobs are retried at most three times.
+
+Operators review jobs that still fail after retries. Queue state is held in
+memory and is not durable across a process restart.
+
+This guarantees zero job loss and means operators never need to monitor the queue.

package/evals/fixtures/i3-documentation-and-adrs/prose-review/queue.js

@@ -0,0 +1,32 @@
+'use strict';
+
+const MAX_BATCH_SIZE = 50;
+const RETENTION_MS = 72 * 60 * 60 * 1000;
+const MAX_RETRIES = 3;
+const jobs = new Map();
+
+function submit(batch, now = Date.now()) {
+  if (batch.length > MAX_BATCH_SIZE) throw new Error('Batch exceeds limit');
+  for (const job of batch) {
+    jobs.set(job.id, { ...job, createdAt: now, retries: 0, state: 'queued' });
+  }
+}
+
+function recordFailure(id) {
+  const job = jobs.get(id);
+  if (!job) throw new Error('Unknown job');
+  if (job.retries < MAX_RETRIES) {
+    job.retries += 1;
+    job.state = 'queued';
+  } else {
+    job.state = 'operator-review';
+  }
+}
+
+function expire(now = Date.now()) {
+  for (const [id, job] of jobs) {
+    if (now - job.createdAt >= RETENTION_MS) jobs.delete(id);
+  }
+}
+
+module.exports = { submit, recordFailure, expire, MAX_BATCH_SIZE, RETENTION_MS, MAX_RETRIES };

package/evals/fixtures/i3-ui-prototype/business-brief.md

@@ -0,0 +1,7 @@
+# 设备采购申请原型
+
+用户明确选择模板模式。比较三套模板，并以企业管理后台模板制作可操作原型。
+申请包含设备名称、申请部门、数量、单价和用途。数量为正整数，单价大于零，计算总金额。
+申请人可编辑自己的草稿、提交申请；审批人可审批待审批记录，退回时必须填写原因；只读人员只能查看。
+覆盖表单校验、提交失败内容保留、审批通过和退回、无权限操作、空列表、键盘焦点与窄屏场景。
+交付需求映射、设计契约、启动方式和实际检查记录。没有浏览器条件时标记待验证。当前没有用户对任何原型版本的认可。

package/evals/plugin/tdd-fires-on-pasted-bug/graders/failing-test-first.md

@@ -0,0 +1,7 @@
+---
+type: llm
+---
+
+PASS if the reply gives a test that reproduces the bug (splitCents(100, 3) summing to 100, or the missing cent) and presents it as the first step, to be run and seen failing before the function is changed, with the fix coming after it.
+
+FAIL if the reply changes the function without a reproducing test, or adds tests only after the fix as an afterthought, or never says the test should fail first.

package/evals/plugin/tdd-fires-on-pasted-bug/graders/skill-fired.md

@@ -0,0 +1,5 @@
+---
+type: tool_used
+tool: Skill
+input_match: '"skill"\s*:\s*"(?:[\w-]+:)?(i3-test-driven-development|test)"'
+---

package/evals/plugin/tdd-fires-on-pasted-bug/prompt.md

@@ -0,0 +1,15 @@
+---
+description: A bug fix request with the function pasted inline and no mention of tests. The skill should fire, and the reply should reproduce the bug with a failing test before fixing it.
+expected_outcome: A test that reproduces the lost cent, shown or described as failing first, then the fix; i3-test-driven-development is invoked.
+max_turns: 12
+allowed_tools: [Read, Glob, Grep, Skill]
+---
+
+Fix this bug: splitCents(100, 3) returns [33, 33, 33] and a cent goes missing. Here is the function.
+
+```js
+function splitCents(totalCents, n) {
+  const share = Math.floor(totalCents / n);
+  return Array.from({ length: n }, () => share);
+}
+```