@fitlab-ai/agent-infra 0.3.2 → 0.4.1

package/templates/.agents/QUICKSTART.md

@@ -8,12 +8,22 @@ This guide walks you through using multiple AI coding assistants together on a p
 - A project with `.agents/` directory set up (this project)
 - Familiarity with your project's codebase
 
+## Git Hook Setup
+
+Enable the shared Git hooks path before relying on the template hook chain:
+
+```bash
+git config core.hooksPath .github/hooks
+```
+
+This makes Git invoke the hooks synced into `.github/hooks/`, including `pre-commit` and `check-version-format.sh`.
+
 ## Creating Your First Task
 
 1. Copy the task template to the active workspace:
 
 ```bash
-cp .agents/templates/task.md .agent-workspace/active/task-001.md
+cp .agents/templates/task.md .agents/workspace/active/task-001.md
 ```
 
 2. Fill in the task metadata:
@@ -49,7 +59,7 @@ Claude Code excels at codebase exploration and understanding complex relationshi
 gemini
 
 # Example prompt:
-# "Based on the analysis in .agent-workspace/active/task-001.md,
+# "Based on the analysis in .agents/workspace/active/task-001.md,
 #  create a technical design. Define interfaces and outline the approach."
 ```
 
@@ -60,7 +70,7 @@ gemini
 codex
 
 # Example prompt:
-# "Implement the changes described in .agent-workspace/active/task-001.md.
+# "Implement the changes described in .agents/workspace/active/task-001.md.
 #  Follow the design section. Create a new branch for this work."
 ```
 
@@ -87,7 +97,7 @@ claude
 
 ```bash
 # Quick bug fix workflow
-cp .agents/templates/task.md .agent-workspace/active/bugfix-001.md
+cp .agents/templates/task.md .agents/workspace/active/bugfix-001.md
 # Edit the task, then:
 # 1. Use Claude Code to analyze
 # 2. Use Codex/Cursor to fix
@@ -101,7 +111,7 @@ cp .agents/templates/task.md .agent-workspace/active/bugfix-001.md
 3. **Report**: Generate review report from template.
 
 ```bash
-cp .agents/templates/review-report.md .agent-workspace/active/review-pr-42.md
+cp .agents/templates/review-report.md .agents/workspace/active/review-pr-42.md
 # Use Claude Code to fill in the review
 ```
 
@@ -113,7 +123,7 @@ cp .agents/templates/review-report.md .agent-workspace/active/review-pr-42.md
 4. **Verify** (Claude Code): Ensure no regressions, run tests.
 
 ```bash
-cp .agents/templates/task.md .agent-workspace/active/refactor-001.md
+cp .agents/templates/task.md .agents/workspace/active/refactor-001.md
 # Follow the refactoring workflow in .agents/workflows/refactoring.yaml
 ```
 
@@ -122,7 +132,7 @@ cp .agents/templates/task.md .agent-workspace/active/refactor-001.md
 When switching between AI tools, create a handoff document:
 
 ```bash
-cp .agents/templates/handoff.md .agent-workspace/active/handoff-task-001-phase2.md
+cp .agents/templates/handoff.md .agents/workspace/active/handoff-task-001-phase2.md
 ```
 
 Fill in:

package/templates/.agents/QUICKSTART.zh-CN.md

@@ -8,12 +8,22 @@
 - 项目已设置 `.agents/` 目录（本项目已就绪）
 - 熟悉你的项目代码库
 
+## Git Hook 配置
+
+在依赖模板中的 Git hook 链路前，先启用共享 hooks 路径：
+
+```bash
+git config core.hooksPath .github/hooks
+```
+
+这样 Git 才会调用同步到 `.github/hooks/` 下的 hook，包括 `pre-commit` 和 `check-version-format.sh`。
+
 ## 创建第一个任务
 
 1. 将任务模板复制到活跃工作区：
 
 ```bash
-cp .agents/templates/task.zh-CN.md .agent-workspace/active/task-001.md
+cp .agents/templates/task.zh-CN.md .agents/workspace/active/task-001.md
 ```
 
 2. 填写任务元数据：
@@ -49,7 +59,7 @@ Claude Code 擅长代码库探索和理解文件间的复杂关系。
 gemini
 
 # 示例提示：
-# "基于 .agent-workspace/active/task-001.md 中的分析结果，
+# "基于 .agents/workspace/active/task-001.md 中的分析结果，
 #  创建技术设计方案。定义接口并概述实现思路。"
 ```
 
@@ -60,7 +70,7 @@ gemini
 codex
 
 # 示例提示：
-# "实现 .agent-workspace/active/task-001.md 中描述的变更。
+# "实现 .agents/workspace/active/task-001.md 中描述的变更。
 #  遵循设计部分。为此工作创建新分支。"
 ```
 
@@ -87,7 +97,7 @@ claude
 
 ```bash
 # 快速缺陷修复工作流
-cp .agents/templates/task.zh-CN.md .agent-workspace/active/bugfix-001.md
+cp .agents/templates/task.zh-CN.md .agents/workspace/active/bugfix-001.md
 # 编辑任务，然后：
 # 1. 使用 Claude Code 分析
 # 2. 使用 Codex/Cursor 修复
@@ -101,7 +111,7 @@ cp .agents/templates/task.zh-CN.md .agent-workspace/active/bugfix-001.md
 3. **报告**：从模板生成审查报告。
 
 ```bash
-cp .agents/templates/review-report.zh-CN.md .agent-workspace/active/review-pr-42.md
+cp .agents/templates/review-report.zh-CN.md .agents/workspace/active/review-pr-42.md
 # 使用 Claude Code 填写审查内容
 ```
 
@@ -113,7 +123,7 @@ cp .agents/templates/review-report.zh-CN.md .agent-workspace/active/review-pr-42
 4. **验证**（Claude Code）：确保没有回归问题，运行测试。
 
 ```bash
-cp .agents/templates/task.zh-CN.md .agent-workspace/active/refactor-001.md
+cp .agents/templates/task.zh-CN.md .agents/workspace/active/refactor-001.md
 # 按照 .agents/workflows/refactoring.zh-CN.yaml 中的重构工作流执行
 ```
 
@@ -122,7 +132,7 @@ cp .agents/templates/task.zh-CN.md .agent-workspace/active/refactor-001.md
 在 AI 工具之间切换时，创建交接文档：
 
 ```bash
-cp .agents/templates/handoff.zh-CN.md .agent-workspace/active/handoff-task-001-phase2.md
+cp .agents/templates/handoff.zh-CN.md .agents/workspace/active/handoff-task-001-phase2.md
 ```
 
 填写：

package/templates/.agents/README.md

@@ -36,12 +36,11 @@ This dual-config approach ensures every AI tool receives appropriate project con
     bug-fix.yaml                # Bug fix workflow
     code-review.yaml            # Code review workflow
     refactoring.yaml            # Refactoring workflow
-
-.agent-workspace/               # Runtime workspace (git-ignored)
-  active/                       # Currently active tasks
-  blocked/                      # Blocked tasks
-  completed/                    # Completed tasks
-  logs/                         # Collaboration logs
+  workspace/                    # Runtime workspace (git-ignored)
+    active/                     # Currently active tasks
+    blocked/                    # Blocked tasks
+    completed/                  # Completed tasks
+    logs/                       # Collaboration logs
 
 .claude/                        # Claude Code specific config
   CLAUDE.md                     # Project instructions for Claude
@@ -53,11 +52,12 @@ This dual-config approach ensures every AI tool receives appropriate project con
 
 The multi-AI collaboration follows a structured workflow:
 
-```
-1. Analysis    -->  2. Design    -->  3. Implementation
-                                            |
-6. Commit  <--  5. Fix Issues  <--  4. Review
-```
+1. Analysis
+2. Design
+3. Implementation
+4. Review
+5. Fix Issues
+6. Commit
 
 ### Phase Details
 
@@ -95,7 +95,7 @@ Each AI tool has different strengths. Use them accordingly:
 ## Quick Start
 
 1. **Read the quick start guide**: See `QUICKSTART.md` for step-by-step instructions.
-2. **Create a task**: Copy `.agents/templates/task.md` to `.agent-workspace/active/`.
+2. **Create a task**: Copy `.agents/templates/task.md` to `.agents/workspace/active/`.
 3. **Assign to an AI**: Update the `assigned_to` field in the task metadata.
 4. **Run the workflow**: Follow the appropriate workflow in `.agents/workflows/`.
 5. **Hand off**: When switching AIs, create a handoff document from the template.
@@ -123,6 +123,13 @@ When writing or updating `.agents/skills/*/SKILL.md` files and their templates,
 5. When renumbering, update every in-document step reference so the instructions remain accurate.
 6. Extract long bash scripts into a sibling `scripts/` directory; the SKILL.md should contain only a single-line invocation (e.g., `bash .agents/skills/<skill>/scripts/<script>.sh`) and a brief summary of the script's responsibilities.
 
+### SKILL.md Size Control
+
+- Keep the SKILL.md body within about 500 tokens (roughly 80 lines / 2KB).
+- Move content beyond that threshold into a sibling `reference/` directory.
+- Use explicit navigation in the skeleton, such as: `Read reference/xxx.md before executing this step.`
+- Keep scripts in `scripts/` and execute them instead of inlining long bash blocks.
+
 ## FAQ
 
 ### Q: Do I need to configure every AI tool separately?
@@ -131,7 +138,7 @@ No. Claude Code reads from `.claude/CLAUDE.md`, and all other tools read from `A
 
 ### Q: How do tasks get passed between AI tools?
 
-Through handoff documents stored in `.agent-workspace/`. Each handoff includes context, progress, and next steps so the receiving AI can continue seamlessly.
+Through handoff documents stored in `.agents/workspace/`. Each handoff includes context, progress, and next steps so the receiving AI can continue seamlessly.
 
 ### Q: What if an AI tool doesn't support AGENTS.md?
 
@@ -143,4 +150,4 @@ It's not recommended. The workflow model is sequential -- one AI per phase. Para
 
 ### Q: Where are runtime files stored?
 
-In `.agent-workspace/`, which is git-ignored. Only templates and workflow definitions in `.agents/` are version-controlled.
+In `.agents/workspace/`, which is git-ignored. Only templates and workflow definitions in `.agents/` are version-controlled.

package/templates/.agents/README.zh-CN.md

@@ -36,12 +36,11 @@
     bug-fix.yaml                # 缺陷修复工作流
     code-review.yaml            # 代码审查工作流
     refactoring.yaml            # 重构工作流
-
-.agent-workspace/               # 运行时工作区（已被 git ignore）
-  active/                       # 当前活跃任务
-  blocked/                      # 被阻塞的任务
-  completed/                    # 已完成的任务
-  logs/                         # 协作日志
+  workspace/                    # 运行时工作区（已被 git ignore）
+    active/                     # 当前活跃任务
+    blocked/                    # 被阻塞的任务
+    completed/                  # 已完成的任务
+    logs/                       # 协作日志
 
 .claude/                        # Claude Code 专属配置
   CLAUDE.md                     # Claude 项目指令
@@ -53,11 +52,12 @@
 
 多 AI 协作遵循结构化工作流：
 
-```
-1. 分析    -->  2. 设计    -->  3. 实现
-                                    |
-6. 提交  <--  5. 修复问题  <--  4. 审查
-```
+1. 分析
+2. 设计
+3. 实现
+4. 审查
+5. 修复问题
+6. 提交
 
 ### 阶段详情
 
@@ -95,7 +95,7 @@
 ## 快速入门
 
 1. **阅读快速入门指南**：参见 `QUICKSTART.zh-CN.md` 获取分步说明。
-2. **创建任务**：将 `.agents/templates/task.zh-CN.md` 复制到 `.agent-workspace/active/`。
+2. **创建任务**：将 `.agents/templates/task.zh-CN.md` 复制到 `.agents/workspace/active/`。
 3. **分配给 AI**：更新任务元数据中的 `assigned_to` 字段。
 4. **执行工作流**：按照 `.agents/workflows/` 中相应的工作流执行。
 5. **交接**：切换 AI 时，从模板创建交接文档。
@@ -123,6 +123,13 @@
 5. 调整编号时，必须同步更新文中的步骤引用，确保说明、命令和检查点一致。
 6. 长 bash 脚本应从 SKILL.md 提取到同级 `scripts/` 目录中，SKILL.md 只保留单行调用（如 `bash .agents/skills/<skill>/scripts/<script>.sh`）和对脚本职责的概要说明。
 
+### SKILL.md 体积控制
+
+- SKILL.md 正文控制在约 500 tokens（约 80 行 / 2KB）以内。
+- 超过阈值的内容拆分到同级 `reference/` 目录。
+- 骨架中使用明确导航，例如：`执行此步骤前，先读取 reference/xxx.md。`
+- 长脚本继续放在 `scripts/` 目录，优先执行脚本而不是内联大段 bash。
+
 ## 常见问题
 
 ### Q：我需要单独配置每个 AI 工具吗？
@@ -131,7 +138,7 @@
 
 ### Q：任务如何在 AI 工具之间传递？
 
-通过存储在 `.agent-workspace/` 中的交接文档。每份交接文档包含上下文、进度和后续步骤，让接收方 AI 能无缝接续。
+通过存储在 `.agents/workspace/` 中的交接文档。每份交接文档包含上下文、进度和后续步骤，让接收方 AI 能无缝接续。
 
 ### Q：如果某个 AI 工具不支持 AGENTS.md 怎么办？
 
@@ -143,4 +150,4 @@
 
 ### Q：运行时文件存储在哪里？
 
-在 `.agent-workspace/` 中，该目录已被 git ignore。只有 `.agents/` 中的模板和工作流定义受版本控制。
+在 `.agents/workspace/` 中，该目录已被 git ignore。只有 `.agents/` 中的模板和工作流定义受版本控制。

package/templates/.agents/skills/analyze-task/SKILL.md

@@ -1,9 +1,6 @@
 ---
 name: analyze-task
-description: >
-  Analyze an existing task and output a requirements analysis document,
-  supporting multi-round artifacts (`analysis.md` / `analysis-r{N}.md`).
-  Triggered when the user asks to analyze a task. Argument: task-id.
+description: "Analyze a task and produce a requirements document"
 ---
 
 # Analyze Task
@@ -19,7 +16,7 @@ description: >
 ### 1. Verify Prerequisites
 
 Check required files:
-- `.agent-workspace/active/{task-id}/task.md` - Task file
+- `.agents/workspace/active/{task-id}/task.md` - Task file
 
 Note: `{task-id}` format is `TASK-{yyyyMMdd-HHmmss}`, for example `TASK-20260306-143022`
 
@@ -27,7 +24,7 @@ If `task.md` is missing, tell the user to create or import the task first.
 
 ### 2. Determine the Analysis Round
 
-Scan `.agent-workspace/active/{task-id}/` for analysis artifact files:
+Scan `.agents/workspace/active/{task-id}/` for analysis artifact files:
 - If neither `analysis.md` nor `analysis-r*.md` exists -> this is Round 1 and must create `analysis.md`
 - If `analysis.md` exists and no `analysis-r*.md` exists -> this is Round 2 and must create `analysis-r2.md`
 - If `analysis-r{N}.md` exists -> this is Round N+1 and must create `analysis-r{N+1}.md`
@@ -61,7 +58,7 @@ Follow the `analysis` step in `.agents/workflows/feature-development.yaml`:
 
 ### 5. Output Analysis Document
 
-Create `.agent-workspace/active/{task-id}/{analysis-artifact}`.
+Create `.agents/workspace/active/{task-id}/{analysis-artifact}`.
 
 ## Output Template
 
@@ -109,7 +106,7 @@ Get the current time:
 date "+%Y-%m-%d %H:%M:%S"
 ```
 
-Update `.agent-workspace/active/{task-id}/task.md`:
+Update `.agents/workspace/active/{task-id}/task.md`:
 - `current_step`: requirement-analysis
 - `assigned_to`: {current AI agent}
 - `updated_at`: {current time}
@@ -135,7 +132,7 @@ Summary:
 - Risk level: {assessment}
 
 Output file:
-- Analysis report: .agent-workspace/active/{task-id}/{analysis-artifact}
+- Analysis report: .agents/workspace/active/{task-id}/{analysis-artifact}
 
 Next step - create technical plan:
   - Claude Code / OpenCode: /plan-task {task-id}
@@ -146,7 +143,7 @@ Next step - create technical plan:
 ## Completion Checklist
 
 - [ ] Read and understood the task file and source information
-- [ ] Created analysis document `.agent-workspace/active/{task-id}/{analysis-artifact}`
+- [ ] Created analysis document `.agents/workspace/active/{task-id}/{analysis-artifact}`
 - [ ] Updated `current_step` to requirement-analysis in task.md
 - [ ] Updated `updated_at` to the current time in task.md
 - [ ] Updated `assigned_to` in task.md

package/templates/.agents/skills/analyze-task/SKILL.zh-CN.md

@@ -1,8 +1,6 @@
 ---
 name: analyze-task
-description: >
-  分析已有任务并输出需求分析文档，支持多轮分析产物（analysis.md / analysis-r{N}.md）。
-  当用户要求分析某个任务时触发。参数：task-id。
+description: "分析任务并输出需求分析文档"
 ---
 
 # 分析任务
@@ -18,7 +16,7 @@ description: >
 ### 1. 验证前置条件
 
 检查必要文件：
-- `.agent-workspace/active/{task-id}/task.md` - 任务文件
+- `.agents/workspace/active/{task-id}/task.md` - 任务文件
 
 注意：`{task-id}` 格式为 `TASK-{yyyyMMdd-HHmmss}`，例如 `TASK-20260306-143022`
 
@@ -26,7 +24,7 @@ description: >
 
 ### 2. 确定分析轮次
 
-扫描 `.agent-workspace/active/{task-id}/` 目录中的分析产物文件：
+扫描 `.agents/workspace/active/{task-id}/` 目录中的分析产物文件：
 - 如果不存在 `analysis.md` 且不存在 `analysis-r*.md` → 本轮为第 1 轮，产出 `analysis.md`
 - 如果存在 `analysis.md` 且不存在 `analysis-r*.md` → 本轮为第 2 轮，产出 `analysis-r2.md`
 - 如果存在 `analysis-r{N}.md` → 本轮为第 N+1 轮，产出 `analysis-r{N+1}.md`
@@ -60,7 +58,7 @@ description: >
 
 ### 5. 输出分析文档
 
-创建 `.agent-workspace/active/{task-id}/{analysis-artifact}`。
+创建 `.agents/workspace/active/{task-id}/{analysis-artifact}`。
 
 ## 输出模板
 
@@ -108,7 +106,7 @@ description: >
 date "+%Y-%m-%d %H:%M:%S"
 ```
 
-更新 `.agent-workspace/active/{task-id}/task.md`：
+更新 `.agents/workspace/active/{task-id}/task.md`：
 - `current_step`：requirement-analysis
 - `assigned_to`：{当前 AI 代理}
 - `updated_at`：{当前时间}
@@ -134,7 +132,7 @@ date "+%Y-%m-%d %H:%M:%S"
 - 风险等级：{评估}
 
 产出文件：
-- 分析报告：.agent-workspace/active/{task-id}/{analysis-artifact}
+- 分析报告：.agents/workspace/active/{task-id}/{analysis-artifact}
 
 下一步 - 设计技术方案：
   - Claude Code / OpenCode：/plan-task {task-id}
@@ -145,7 +143,7 @@ date "+%Y-%m-%d %H:%M:%S"
 ## 完成检查清单
 
 - [ ] 阅读并理解了任务文件和来源信息
-- [ ] 创建了分析文档 `.agent-workspace/active/{task-id}/{analysis-artifact}`
+- [ ] 创建了分析文档 `.agents/workspace/active/{task-id}/{analysis-artifact}`
 - [ ] 更新了 task.md 中的 `current_step` 为 requirement-analysis
 - [ ] 更新了 task.md 中的 `updated_at` 为当前时间
 - [ ] 更新了 task.md 中的 `assigned_to`

package/templates/.agents/skills/block-task/SKILL.md

@@ -1,10 +1,6 @@
 ---
 name: block-task
-description: >
-  Mark a task as blocked and record the blocking reason, moving it from
-  the active directory to the blocked directory. Use when a task cannot
-  proceed due to technical problems, unclear requirements, missing resources,
-  or pending decisions. Arguments: task-id, optional blocking reason.
+description: "Mark a task as blocked and record the reason"
 ---
 
 # Block Task
@@ -25,7 +21,7 @@ description: >
 
 ### 1. Verify Task Exists
 
-Check that the task exists in `.agent-workspace/active/{task-id}/`.
+Check that the task exists in `.agents/workspace/active/{task-id}/`.
 
 Note: `{task-id}` format is `TASK-{yyyyMMdd-HHmmss}`, e.g. `TASK-20260306-143022`
 
@@ -47,7 +43,7 @@ Get the current time:
 date "+%Y-%m-%d %H:%M:%S"
 ```
 
-Update `.agent-workspace/active/{task-id}/task.md`:
+Update `.agents/workspace/active/{task-id}/task.md`:
 - `status`: blocked
 - `blocked_at`: {current timestamp}
 - `updated_at`: {current timestamp}
@@ -61,13 +57,13 @@ Add a blocking information section to task.md.
 ### 4. Move Task to Blocked Directory
 
 ```bash
-mv .agent-workspace/active/{task-id} .agent-workspace/blocked/{task-id}
+mv .agents/workspace/active/{task-id} .agents/workspace/blocked/{task-id}
 ```
 
 ### 5. Verify Move
 
 ```bash
-ls .agent-workspace/blocked/{task-id}/task.md
+ls .agents/workspace/blocked/{task-id}/task.md
 ```
 
 ### 6. Sync to Issue (Optional)
@@ -93,10 +89,10 @@ Task {task-id} marked as blocked.
 
 Blocking reason: {summary}
 Required to unblock: {what's needed}
-Archived to: .agent-workspace/blocked/{task-id}/
+Archived to: .agents/workspace/blocked/{task-id}/
 
 To unblock when the issue is resolved:
-  mv .agent-workspace/blocked/{task-id} .agent-workspace/active/{task-id}
+  mv .agents/workspace/blocked/{task-id} .agents/workspace/active/{task-id}
   # Then update task.md: status -> active, remove blocked_at
 ```
 
@@ -133,7 +129,7 @@ Blocking information section to add to task.md:
 
 - [ ] Analyzed and documented the blocking reason
 - [ ] Updated task.md with blocked status and blocking information
-- [ ] Moved task directory to `.agent-workspace/blocked/`
+- [ ] Moved task directory to `.agents/workspace/blocked/`
 - [ ] Verified move succeeded
 - [ ] Informed user how to unblock
 
@@ -143,7 +139,7 @@ When the blocking issue is resolved:
 
 ```bash
 # 1. Move back to active
-mv .agent-workspace/blocked/{task-id} .agent-workspace/active/{task-id}
+mv .agents/workspace/blocked/{task-id} .agents/workspace/active/{task-id}
 
 # 2. Update task.md: set status to active, update timestamps
 # 3. Resume from where you left off (check current_step)

package/templates/.agents/skills/block-task/SKILL.zh-CN.md

@@ -1,9 +1,6 @@
 ---
 name: block-task
-description: >
-  标记任务为阻塞状态并记录阻塞原因，将其从 active 目录移动到 blocked 目录。
-  当任务因技术问题、需求不清晰、缺少资源或待定决策而无法继续时使用。
-  参数：task-id，可选的阻塞原因。
+description: "标记任务为阻塞状态并记录原因"
 ---
 
 # 标记任务阻塞
@@ -24,7 +21,7 @@ description: >
 
 ### 1. 验证任务存在
 
-检查任务是否存在于 `.agent-workspace/active/{task-id}/`。
+检查任务是否存在于 `.agents/workspace/active/{task-id}/`。
 
 注意：`{task-id}` 格式为 `TASK-{yyyyMMdd-HHmmss}`，例如 `TASK-20260306-143022`
 
@@ -46,7 +43,7 @@ description: >
 date "+%Y-%m-%d %H:%M:%S"
 ```
 
-更新 `.agent-workspace/active/{task-id}/task.md`：
+更新 `.agents/workspace/active/{task-id}/task.md`：
 - `status`：blocked
 - `blocked_at`：{当前时间戳}
 - `updated_at`：{当前时间戳}
@@ -60,13 +57,13 @@ date "+%Y-%m-%d %H:%M:%S"
 ### 4. 移动任务到 blocked 目录
 
 ```bash
-mv .agent-workspace/active/{task-id} .agent-workspace/blocked/{task-id}
+mv .agents/workspace/active/{task-id} .agents/workspace/blocked/{task-id}
 ```
 
 ### 5. 验证移动
 
 ```bash
-ls .agent-workspace/blocked/{task-id}/task.md
+ls .agents/workspace/blocked/{task-id}/task.md
 ```
 
 ### 6. 同步到 Issue（可选）
@@ -92,10 +89,10 @@ ls .agent-workspace/blocked/{task-id}/task.md
 
 阻塞原因：{摘要}
 解除阻塞所需：{需要什么}
-归档路径：.agent-workspace/blocked/{task-id}/
+归档路径：.agents/workspace/blocked/{task-id}/
 
 解除阻塞时执行：
-  mv .agent-workspace/blocked/{task-id} .agent-workspace/active/{task-id}
+  mv .agents/workspace/blocked/{task-id} .agents/workspace/active/{task-id}
   # 然后更新 task.md：status -> active，移除 blocked_at
 ```
 
@@ -132,7 +129,7 @@ ls .agent-workspace/blocked/{task-id}/task.md
 
 - [ ] 分析并记录了阻塞原因
 - [ ] 更新了 task.md 的阻塞状态和阻塞信息
-- [ ] 将任务目录移动到 `.agent-workspace/blocked/`
+- [ ] 将任务目录移动到 `.agents/workspace/blocked/`
 - [ ] 验证了移动成功
 - [ ] 告知了用户如何解除阻塞
 
@@ -142,7 +139,7 @@ ls .agent-workspace/blocked/{task-id}/task.md
 
 ```bash
 # 1. 移回 active
-mv .agent-workspace/blocked/{task-id} .agent-workspace/active/{task-id}
+mv .agents/workspace/blocked/{task-id} .agents/workspace/active/{task-id}
 
 # 2. 更新 task.md：设置 status 为 active，更新时间戳
 # 3. 从中断处继续（检查 current_step）

package/templates/.agents/skills/check-task/SKILL.md

@@ -1,10 +1,6 @@
 ---
 name: check-task
-description: >
-  Check a task's current status, workflow progress, and context files. This is a
-  read-only operation that reports the task status and recommends appropriate
-  next steps. Triggered when the user asks to inspect task status. Argument:
-  task-id.
+description: "Check a task's current status and progress"
 ---
 
 # Check Task Status
@@ -19,9 +15,9 @@ description: >
 ### 1. Locate Task
 
 Search for the task in this priority order:
-1. `.agent-workspace/active/{task-id}/task.md`
-2. `.agent-workspace/blocked/{task-id}/task.md`
-3. `.agent-workspace/completed/{task-id}/task.md`
+1. `.agents/workspace/active/{task-id}/task.md`
+2. `.agents/workspace/blocked/{task-id}/task.md`
+3. `.agents/workspace/completed/{task-id}/task.md`
 
 Note: `{task-id}` format is `TASK-{yyyyMMdd-HHmmss}`, for example `TASK-20260306-143022`
 

package/templates/.agents/skills/check-task/SKILL.zh-CN.md

@@ -1,8 +1,6 @@
 ---
 name: check-task
-description: >
-  查看任务的当前状态、工作流进度和上下文文件。这是只读操作，报告任务状态并建议
-  适当的下一步操作。当用户要求查看任务状态时触发。参数：task-id。
+description: "查看任务的当前状态和进度"
 ---
 
 # 查看任务状态
@@ -17,9 +15,9 @@ description: >
 ### 1. 查找任务
 
 按以下优先顺序搜索任务：
-1. `.agent-workspace/active/{task-id}/task.md`
-2. `.agent-workspace/blocked/{task-id}/task.md`
-3. `.agent-workspace/completed/{task-id}/task.md`
+1. `.agents/workspace/active/{task-id}/task.md`
+2. `.agents/workspace/blocked/{task-id}/task.md`
+3. `.agents/workspace/completed/{task-id}/task.md`
 
 注意：`{task-id}` 格式为 `TASK-{yyyyMMdd-HHmmss}`，例如 `TASK-20260306-143022`
 

package/templates/.agents/skills/close-codescan/SKILL.md

@@ -1,9 +1,6 @@
 ---
 name: close-codescan
-description: >
-  Dismiss a Code Scanning (CodeQL) alert with a documented justification.
-  Triggered when the user asks to dismiss a Code Scanning alert.
-  Argument: alert number.
+description: "Close a Code Scanning alert with a documented reason"
 ---
 
 # Dismiss Code Scanning Alert

package/templates/.agents/skills/close-codescan/SKILL.zh-CN.md

@@ -1,8 +1,6 @@
 ---
 name: close-codescan
-description: >
-  关闭 Code Scanning（CodeQL）告警并提供有据可查的理由。
-  当用户要求关闭 Code Scanning 告警时触发。参数：告警编号。
+description: "关闭 Code Scanning 告警并记录理由"
 ---
 
 # 关闭 Code Scanning 告警

package/templates/.agents/skills/close-dependabot/SKILL.md

@@ -1,9 +1,6 @@
 ---
 name: close-dependabot
-description: >
-  Dismiss a Dependabot security alert with a documented justification.
-  Triggered when the user asks to dismiss a Dependabot alert.
-  Argument: alert number.
+description: "Close a Dependabot alert with a documented reason"
 ---
 
 # Dismiss Dependabot Alert

package/templates/.agents/skills/close-dependabot/SKILL.zh-CN.md

@@ -1,8 +1,6 @@
 ---
 name: close-dependabot
-description: >
-  关闭 Dependabot 安全告警并提供有据可查的理由。
-  当用户要求关闭 Dependabot 告警时触发。参数：告警编号。
+description: "关闭 Dependabot 安全告警并记录理由"
 ---
 
 # 关闭 Dependabot 告警