@tinkcarlos/skillora 0.2.0

package/.claude/skills/verification-before-completion/SKILL.md

@@ -0,0 +1,305 @@
+---
+name: verification-before-completion
+description: |
+  Ensure work is actually complete before declaring done.
+
+  Use when:
+  - About to declare a task/feature complete
+  - After all implementation tasks are done
+  - Before creating PR or merging
+
+  Key Features:
+  - Comprehensive verification checklist
+  - Evidence-based completion (not assumptions)
+  - Prevents premature "done" declarations
+
+  Core Principle: "Verify before declaring success"
+allowed-tools: [read, write, execute, grep, glob]
+---
+
+# Verification Before Completion
+
+**Core Principle**: Verify before declaring success. Evidence over claims.
+
+## When to Use
+
+- After completing all implementation tasks
+- Before declaring any work "done"
+- Before creating PR or merging
+- When tempted to say "it should work"
+
+**Announce at start:** "I'm using the verification-before-completion skill to ensure this is actually done."
+
+## The Iron Law
+
+```
+NEVER DECLARE DONE WITHOUT VERIFICATION
+```
+
+"It should work" is not verification.
+"I think it's done" is not verification.
+"The code looks right" is not verification.
+
+**Only passing tests and working functionality count.**
+
+## 🚪 5-Step Gate Function (MANDATORY)
+
+```
+IDENTIFY → RUN → READ → VERIFY → CLAIM
+```
+
+| Step | Action | Output |
+|------|--------|--------|
+| **IDENTIFY** | 确定需要验证什么 | 验证清单 |
+| **RUN** | 执行验证命令 | 命令输出 |
+| **READ** | 阅读并理解输出 | 输出解读 |
+| **VERIFY** | 确认输出符合预期 | 通过/失败 |
+| **CLAIM** | 只有全部通过才能声称完成 | 完成声明 |
+
+**每一步都必须有证据，不能跳过！**
+
+## 🚩 Red Flag Words Detection
+
+**以下词汇表示未完成验证 - 如果你的报告包含这些词，必须停止并重新验证:**
+
+| 红旗词 | 问题 |
+|-------|------|
+| "应该" (should) | 未实际验证 |
+| "可能" (probably) | 不确定 |
+| "似乎" (seems) | 未确认 |
+| "我认为" (I think) | 主观判断 |
+| "理论上" (theoretically) | 未实践 |
+| "看起来" (looks like) | 未验证 |
+
+**如果你的报告包含这些词:**
+1. 停止声称完成
+2. 执行实际验证
+3. 用证据替换假设
+
+## Verification Checklist
+
+### 1. All Tests Pass
+
+```bash
+# Run full test suite
+npm test          # Node.js
+pytest            # Python
+cargo test        # Rust
+
+# Expected: ALL TESTS PASS
+# No skipped tests
+# No warnings
+```
+
+**If any test fails:** NOT DONE. Fix first.
+
+### 2. No Lint Errors
+
+```bash
+# Check linting
+npm run lint      # Node.js
+flake8 .          # Python
+cargo clippy      # Rust
+
+# Expected: No errors, no warnings
+```
+
+**If lint errors exist:** NOT DONE. Fix first.
+
+### 3. No Type Errors
+
+```bash
+# Check types
+npx tsc --noEmit  # TypeScript
+mypy .            # Python
+cargo check       # Rust
+
+# Expected: No type errors
+```
+
+**If type errors exist:** NOT DONE. Fix first.
+
+### 4. Feature Actually Works
+
+**Manual verification required:**
+
+1. Start the application
+2. Navigate to the feature
+3. Test the happy path
+4. Test edge cases
+5. Test error cases
+
+```markdown
+## Manual Verification
+
+| Scenario | Expected | Actual | Pass? |
+|----------|----------|--------|-------|
+| Happy path | [Expected] | [Actual] | ✅/❌ |
+| Edge case 1 | [Expected] | [Actual] | ✅/❌ |
+| Error case | [Expected] | [Actual] | ✅/❌ |
+```
+
+**If any scenario fails:** NOT DONE. Fix first.
+
+### 5. No Console Errors
+
+```bash
+# Check browser console (web apps)
+# Check terminal output (CLI apps)
+# Check logs (backend apps)
+
+# Expected: No errors, no warnings
+```
+
+**If console errors exist:** NOT DONE. Fix first.
+
+### 6. Documentation Updated
+
+If the feature requires documentation:
+
+- [ ] README updated (if needed)
+- [ ] API docs updated (if needed)
+- [ ] Comments added for complex logic
+- [ ] CHANGELOG updated (if needed)
+
+### 7. Code Committed
+
+```bash
+# Check git status
+git status
+
+# Expected: Nothing to commit, working tree clean
+```
+
+**If uncommitted changes exist:** Commit first.
+
+## Verification Report Template
+
+```markdown
+## Verification Report
+
+**Feature:** [Feature name]
+**Date:** YYYY-MM-DD
+
+### Automated Checks
+
+| Check | Status | Evidence |
+|-------|--------|----------|
+| Tests pass | ✅/❌ | `npm test` output |
+| Lint clean | ✅/❌ | `npm run lint` output |
+| Types clean | ✅/❌ | `tsc --noEmit` output |
+
+### Manual Verification
+
+| Scenario | Expected | Actual | Pass? |
+|----------|----------|--------|-------|
+| [Scenario 1] | [Expected] | [Actual] | ✅/❌ |
+| [Scenario 2] | [Expected] | [Actual] | ✅/❌ |
+
+### Console/Logs
+
+- [ ] No errors in browser console
+- [ ] No errors in server logs
+- [ ] No warnings
+
+### Documentation
+
+- [ ] README updated (if needed)
+- [ ] API docs updated (if needed)
+
+### Git Status
+
+- [ ] All changes committed
+- [ ] Commit messages clear
+
+### Final Status
+
+**VERIFIED COMPLETE:** ✅/❌
+
+**If not complete, remaining issues:**
+- [Issue 1]
+- [Issue 2]
+```
+
+## Common Traps to Avoid
+
+| Trap | Reality |
+|------|---------|
+| "Tests pass locally" | CI might fail |
+| "It works on my machine" | Environment differences |
+| "Just a small change" | Small changes break things |
+| "I'll fix that later" | Later never comes |
+| "It's probably fine" | Probably isn't verification |
+
+## When Verification Fails
+
+If any check fails:
+
+1. **Do NOT declare done**
+2. **Fix the issue**
+3. **Re-run verification**
+4. **Repeat until all pass**
+
+```markdown
+## Verification Failed
+
+**Failed Check:** [Which check failed]
+**Reason:** [Why it failed]
+**Fix Required:** [What needs to be done]
+
+**Status:** NOT COMPLETE - Fixing issues
+```
+
+## Integration with Other Skills
+
+| After | Use |
+|-------|-----|
+| `/subagent-driven-development` | All tasks complete |
+| **This skill** | Verify everything works |
+| `/finishing-a-development-branch` | Merge/PR |
+
+## Output Contract
+
+```markdown
+## Verification Complete
+
+**Feature:** [Feature name]
+
+**All Checks Passed:**
+- [x] Tests pass
+- [x] Lint clean
+- [x] Types clean
+- [x] Manual verification passed
+- [x] No console errors
+- [x] Documentation updated
+- [x] Code committed
+
+**Evidence:**
+- Test output: [summary]
+- Manual test: [summary]
+
+**Status:** ✅ VERIFIED COMPLETE
+
+**Ready for:** `/finishing-a-development-branch`
+```
+
+## Quick Verification Commands
+
+```bash
+# Node.js project
+npm test && npm run lint && npx tsc --noEmit && echo "✅ All checks pass"
+
+# Python project
+pytest && flake8 . && mypy . && echo "✅ All checks pass"
+
+# Rust project
+cargo test && cargo clippy && cargo check && echo "✅ All checks pass"
+```
+
+## Remember
+
+- **Evidence over claims**
+- **Verify, don't assume**
+- **All checks must pass**
+- **Manual verification required**
+- **"Done" means verified done**

package/.claude/skills/writing-plans/SKILL.md

@@ -0,0 +1,259 @@
+---
+name: writing-plans
+description: |
+  Create detailed implementation plans before touching code.
+
+  Use when:
+  - You have a spec or requirements for a multi-step task
+  - Before starting any feature implementation
+  - After brainstorming/design approval
+
+  Key Features:
+  - Bite-sized tasks (2-5 minutes each)
+  - Exact file paths and complete code
+  - TDD-first approach in every task
+  - Frequent commits
+
+  Output: `docs/plans/YYYY-MM-DD-<feature-name>.md`
+allowed-tools: [read, write, execute, grep, glob]
+---
+
+# Writing Plans
+
+**Core Principle**: Write comprehensive implementation plans assuming the engineer has zero context and questionable taste.
+
+## When to Use
+
+- After design approval from brainstorming
+- Before touching any code
+- When you have clear requirements
+
+**Announce at start:** "I'm using the writing-plans skill to create the implementation plan."
+
+## Plan Document Location
+
+Save plans to: `docs/plans/YYYY-MM-DD-<feature-name>.md`
+
+Example: `docs/plans/2025-01-16-user-authentication.md`
+
+## Plan Document Header
+
+**Every plan MUST start with this header:**
+
+```markdown
+# [Feature Name] Implementation Plan
+
+> **For Claude:** REQUIRED SUB-SKILL: Use `executing-plans` or `subagent-driven-development` to implement this plan task-by-task.
+
+**Goal:** [One sentence describing what this builds]
+
+**Architecture:** [2-3 sentences about approach]
+
+**Tech Stack:** [Key technologies/libraries]
+
+**Prerequisites:**
+- [ ] Design document approved
+- [ ] Git worktree created (if applicable)
+- [ ] Dependencies identified
+
+---
+```
+
+## Bite-Sized Task Granularity
+
+**Each step is one action (2-5 minutes):**
+
+| Action Type | Example |
+|-------------|---------|
+| Write test | "Write the failing test for user login" |
+| Verify fail | "Run test to verify it fails" |
+| Implement | "Write minimal code to pass the test" |
+| Verify pass | "Run tests to verify they pass" |
+| Commit | "Commit the changes" |
+
+**NOT bite-sized:**
+- "Implement user authentication" (too big)
+- "Add login functionality" (too vague)
+
+## Task Structure Template
+
+```markdown
+### Task N: [Component Name]
+
+**Files:**
+- Create: `exact/path/to/file.py`
+- Modify: `exact/path/to/existing.py:123-145`
+- Test: `tests/exact/path/to/test.py`
+
+**Step 1: Write the failing test**
+
+\`\`\`python
+def test_specific_behavior():
+    # Arrange
+    input_data = {...}
+
+    # Act
+    result = function(input_data)
+
+    # Assert
+    assert result == expected
+\`\`\`
+
+**Step 2: Run test to verify it fails**
+
+Run: `pytest tests/path/test.py::test_name -v`
+Expected: FAIL with "function not defined"
+
+**Step 3: Write minimal implementation**
+
+\`\`\`python
+def function(input_data):
+    # Minimal implementation
+    return expected
+\`\`\`
+
+**Step 4: Run test to verify it passes**
+
+Run: `pytest tests/path/test.py::test_name -v`
+Expected: PASS
+
+**Step 5: Commit**
+
+\`\`\`bash
+git add tests/path/test.py src/path/file.py
+git commit -m "feat: add specific feature"
+\`\`\`
+```
+
+## Plan Quality Checklist
+
+Every task must have:
+
+- [ ] **Exact file paths** - No "somewhere in src/"
+- [ ] **Complete code** - Not "add validation logic"
+- [ ] **Exact commands** - With expected output
+- [ ] **TDD structure** - Test first, then implement
+- [ ] **Commit step** - After each green
+
+## Key Principles
+
+| Principle | Meaning |
+|-----------|---------|
+| **DRY** | Don't Repeat Yourself |
+| **YAGNI** | You Aren't Gonna Need It |
+| **TDD** | Test-Driven Development |
+| **Frequent Commits** | After each passing test |
+
+## Execution Handoff
+
+After saving the plan, offer execution choice:
+
+```markdown
+**Plan complete and saved to `docs/plans/<filename>.md`.**
+
+**Two execution options:**
+
+1. **Subagent-Driven (this session)**
+   - Fresh subagent per task
+   - Review between tasks
+   - Fast iteration
+   - Use: `/subagent-driven-development`
+
+2. **Batch Execution (with checkpoints)**
+   - Execute in batches
+   - Human checkpoints
+   - Use: `/executing-plans`
+
+**Which approach?**
+```
+
+## Example Plan Snippet
+
+```markdown
+# User Login Implementation Plan
+
+> **For Claude:** REQUIRED SUB-SKILL: Use `subagent-driven-development` to implement.
+
+**Goal:** Allow users to log in with email and password
+
+**Architecture:** JWT-based authentication with bcrypt password hashing
+
+**Tech Stack:** FastAPI, SQLAlchemy, PyJWT, bcryptjs
+
+---
+
+### Task 1: Create User Model Test
+
+**Files:**
+- Create: `tests/models/test_user.py`
+- Create: `src/models/user.py`
+
+**Step 1: Write the failing test**
+
+\`\`\`python
+def test_user_model_has_email_and_password():
+    user = User(email="test@example.com", password="hashed")
+    assert user.email == "test@example.com"
+    assert user.password == "hashed"
+\`\`\`
+
+**Step 2: Run test to verify it fails**
+
+Run: `pytest tests/models/test_user.py -v`
+Expected: FAIL with "cannot import name 'User'"
+
+**Step 3: Write minimal implementation**
+
+\`\`\`python
+from sqlalchemy import Column, String
+from src.database import Base
+
+class User(Base):
+    __tablename__ = "users"
+    email = Column(String, primary_key=True)
+    password = Column(String)
+\`\`\`
+
+**Step 4: Run test to verify it passes**
+
+Run: `pytest tests/models/test_user.py -v`
+Expected: PASS
+
+**Step 5: Commit**
+
+\`\`\`bash
+git add tests/models/test_user.py src/models/user.py
+git commit -m "feat: add User model with email and password"
+\`\`\`
+
+---
+
+### Task 2: Create Login Endpoint Test
+...
+```
+
+## Output Contract
+
+```markdown
+## Plan Created
+
+**File:** `docs/plans/YYYY-MM-DD-<feature>.md`
+
+**Summary:**
+- Total tasks: N
+- Estimated time: X minutes (N tasks × 2-5 min)
+- Files to create: [list]
+- Files to modify: [list]
+
+**Ready for execution?**
+- [ ] Plan reviewed
+- [ ] Execution method chosen
+```
+
+## Remember
+
+- **Exact file paths always**
+- **Complete code in plan** (not "add validation")
+- **Exact commands with expected output**
+- **DRY, YAGNI, TDD, frequent commits**
+- **Assume engineer has zero context**

package/README.md

@@ -0,0 +1,69 @@
+# Skillora
+
+多平台 AI 代理技能管理工具，支持 Claude Code、Cursor、Codex/GPT、Gemini、Windsurf、Aider 等平台。
+
+## 安装
+
+```bash
+npm i -g @tinkcarlos/skillora
+```
+
+> 需要 Node.js 20.6+
+
+## 快速开始
+
+### 交互式初始化（推荐）
+
+```bash
+skillora init
+```
+
+交互式选择平台和安装模式。
+
+### 非交互式初始化
+
+```bash
+# 仅 Claude Code
+skillora init --platform claude-code
+
+# 多平台
+skillora init --platform claude-code,cursor
+
+# 指定安装模式
+skillora init --platform claude-code --universal  # 通用模式
+skillora init --platform claude-code --global     # 全局模式
+```
+
+## 命令参考
+
+| 命令 | 说明 |
+|------|------|
+| `skillora init` | 交互式初始化项目 |
+| `skillora install` | 安装内置技能 |
+| `skillora list` | 列出已安装技能 |
+| `skillora read <name>` | 读取技能内容（给 AI agent 用） |
+| `skillora sync` | 同步配置到 AGENTS.md |
+| `skillora remove <name>` | 删除技能 |
+
+## 支持的平台
+
+| 平台 | 配置文件 |
+|------|----------|
+| Claude Code | `.claude/skills/`, `CLAUDE.md` |
+| Cursor | `.cursorrules`, `AGENTS.md` |
+| Codex/GPT | `AGENTS.md` (System Prompt) |
+| Gemini | `AGENTS.md` (System Prompt) |
+| Windsurf | `AGENTS.md` |
+| Aider | `AGENTS.md` |
+
+## 安装模式
+
+| 模式 | 目录 | 说明 |
+|------|------|------|
+| 项目模式（默认） | `.claude/skills/` | 仅当前项目 |
+| 通用模式 | `.agent/skills/` | 多代理兼容 |
+| 全局模式 | `~/.claude/skills/` | 所有项目共享 |
+
+## 许可证
+
+UNLICENSED