@pixelcraft-tw/spec 1.0.0 → 1.0.1

package/README.md

@@ -44,8 +44,6 @@ Initialize `.workflow/` directory and Claude Code slash commands for your projec
 | Option | Description |
 |--------|-------------|
 | `--force` | Overwrite existing `.workflow/` / 覆蓋現有目錄 |
-| `--arch <pattern>` | Architecture pattern: `clean` \| `hexagonal` \| `ddd` \| `layered` \| `modular` \| `custom` |
-| `--lang <lang>` | Language-framework override (e.g. `go`, `typescript`, `go-gin`) |
 
 ### `pxs new <name>`
 
@@ -168,14 +166,14 @@ After `pxs init`, the following slash commands are available in Claude Code:
 
 | Slash Command | Description |
 |---------------|-------------|
-| `/pxs:new` | Create a new spec / 建立規格 |
-| `/pxs:refine` | Refine spec into plan / 拆解計畫 |
-| `/pxs:clarify` | Clarify requirements / 釐清需求 |
-| `/pxs:implement` | Implement task by task / 逐步實作 |
-| `/pxs:review` | View reviews / 查看 review |
-| `/pxs:status` | Check status / 查看狀態 |
-| `/pxs:reset` | Reset feature phase / 重置 feature 狀態 |
-| `/pxs:diff` | Show task progress / 查看 task 進度 |
+| `/pxs.new` | Create a new spec / 建立規格 |
+| `/pxs.refine` | Refine spec into plan / 拆解計畫 |
+| `/pxs.clarify` | Clarify requirements / 釐清需求 |
+| `/pxs.implement` | Implement task by task / 逐步實作 |
+| `/pxs.review` | View reviews / 查看 review |
+| `/pxs.status` | Check status / 查看狀態 |
+| `/pxs.reset` | Reset feature phase / 重置 feature 狀態 |
+| `/pxs.diff` | Show task progress / 查看 task 進度 |
 
 ## Project Structure / 專案結構
 
@@ -183,7 +181,6 @@ After `pxs init`, the following slash commands are available in Claude Code:
 .workflow/
 ├── state.yaml          # Workflow state / 工作流狀態
 ├── config.yaml         # Project config / 專案設定
-├── architecture.md     # Architecture constraints (if --arch) / 架構約束
 ├── specs/              # Feature specs / 功能規格
 ├── plans/              # Implementation plans / 實作計畫
 ├── reviews/            # AI review records / AI review 紀錄

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pixelcraft-tw/spec",
-  "version": "1.0.0",
+  "version": "1.0.1",
   "description": "Lightweight spec-driven development CLI",
   "type": "module",
   "license": "MIT",

package/templates/claude-commands/sf.clarify.md

@@ -7,12 +7,24 @@ Parse $ARGUMENTS:
 - @ prefixed = agent, / prefixed = skill
 - Remaining = supplementary instructions
 
+<HARD-GATE>
+- This command is STRICTLY for requirement clarification
+- DO NOT write, create, modify, or delete any source code files
+- DO NOT implement any functionality — no application code, no tests, no config changes
+- The ONLY files you may write to:
+  - `.workflow/specs/<name>.md` (updated spec with clarified requirements)
+  - `.workflow/state.yaml` (state updates)
+- Implementation belongs to `/pxs.implement` — not here
+</HARD-GATE>
+
 Read `.workflow/specs/<name>.md`.
 
 Analyze spec, identify requirements that can be further clarified.
 Present as numbered questions, wait for user answers.
 Update spec based on answers.
 
+Remember: Do NOT write any implementation code. Only update the spec document.
+
 {If @agent present, involve corresponding agent}
 {If /skill present, use corresponding MCP/skill to assist}
 {Supplementary instructions}

package/templates/claude-commands/sf.new.md

@@ -7,6 +7,7 @@ Create a spec based on user input.
 Parse $ARGUMENTS:
 - First word is the feature name <name> (MUST be English, kebab-case. If user provides Chinese, translate to English)
 - If `--jira <tickets...>` present, read each ticket's content via Jira MCP (supports multiple, space-separated)
+- If text contains Jira URLs (e.g., `https://xxx.atlassian.net/browse/PROJ-123`), extract ticket IDs and treat as `--jira`
 - If `--desc` present, following text is the description
 - If `-i` present, enter interactive Q&A
 - Remaining text is treated as feature description
@@ -16,7 +17,7 @@ Steps:
 2. Generate spec based on input method:
    - No extra args: copy `.workflow/templates/spec-template.md` to `.workflow/specs/<name>.md`
    - Has description text: expand into complete spec
-   - Has --jira: read Jira content and convert to spec
+   - Has --jira or Jira URLs: read Jira content via MCP and convert to spec
    - Has -i: ask questions step by step to generate spec
 3. Display generated spec content
-4. Prompt user they can manually edit then run `/pxs:refine <name>`
+4. Prompt user they can manually edit then run `/pxs.refine <name>`

package/templates/claude-commands/sf.refine.md

@@ -9,6 +9,17 @@ Parse $ARGUMENTS:
 - --skip-clarify skips requirement clarification
 - Remaining text is supplementary instructions
 
+<HARD-GATE>
+- This command is STRICTLY for spec refinement and plan decomposition
+- DO NOT write, create, modify, or delete any source code files
+- DO NOT implement any functionality — no application code, no tests, no config changes
+- The ONLY files you may write to:
+  - `.workflow/specs/<name>.md` (refined spec)
+  - `.workflow/plans/<name>.md` (decomposed plan)
+  - `.workflow/state.yaml` (state updates)
+- Implementation belongs to `/pxs.implement` — not here
+</HARD-GATE>
+
 Read `.workflow/specs/<name>.md`.
 
 ## Phase 1: Requirement Clarification
@@ -32,6 +43,8 @@ Unless --skip-clarify is present:
 7. Present refined spec to user for confirmation
 8. User approves to continue; otherwise modify based on feedback
 
+Remember: Do NOT write any implementation code. Only update the spec document.
+
 ## Phase 3: Decompose Plan
 1. Determine feature type (feat/fix/refactor/docs/chore)
 2. Decompose requirements into independently completable tasks
@@ -44,4 +57,6 @@ Unless --skip-clarify is present:
 7. User edit → wait for modifications
 8. User re-split → re-decompose based on feedback
 
+Remember: Do NOT write any implementation code. Only generate the plan document.
+
 {Supplementary instructions: $REMAINING_TEXT}