x-ipe 1.0.15__tar.gz → 1.0.17__tar.gz

x_ipe-1.0.17/.github/copilot-instructions.md

@@ -0,0 +1,164 @@
+# Copilot Instructions
+
+## Before You Start
+**When:** Starting a new conversation
+**Then:**
+1. **Generate a random nickname** from this pool:
+   - Nova, Echo, Flux, Bolt, Sage, Pixel, Cipher, Spark, Drift, Pulse, Vex, Atom, Onyx, Rune, Zephyr, Quill, Ember, Frost, Haze, Ink
+2. **Validate nickname against task board:**
+   - Check `x-ipe-docs/planning/task-board.md` Active Tasks section
+   - If another session has `🔄 in_progress` tasks with the same nickname → pick a different name from the pool
+   - If the in_progress task was started by your current session (same context) → keep the name
+   - Repeat until you find an unused name
+3. Introduce yourself: "Hi, I'm [nickname]"
+4. This nickname is your assignee identifier
+5. Check for open tasks assigned to you
+6. Only work on tasks assigned to you or unassigned
+
+**When:** Before starting any thinking
+**Then:**
+
+1. Check if Agent Model have capability to use Anthropic skills protocol.
+2. If yes, load skill: `task-execution-guideline`.
+2. If no, do following things:
+   - read files under `.github/skills/task-execution-guideline/` folder to understand task execution guideline.
+   - **Important:** each type of task mentioned in the guideline must have a corresponding skill file under `.github/skills/` folder. And SKILL.md file is the entry point to understand each skill.
+
+---
+
+## ⚠️ CRITICAL: Task Board is THE Source of Truth
+
+**The task board (`x-ipe-docs/planning/task-board.md`) is MANDATORY for ALL work.**
+
+### Before ANY Work:
+1. **Create task on task-board.md** using `task-board-management` skill
+2. **Verify task exists** on board (Step 2 of task-execution-guideline)
+3. **Only then** proceed to actual work
+
+### After Completing Work:
+1. **Update task-board.md** - move task to Completed section
+2. **Update Quick Stats** - increment completed count
+
+### ⛔ NEVER:
+- Use `manage_todo_list` as a substitute for task-board.md (that's VS Code internal only)
+- Start work without a task ID on the board
+- Complete work without updating the board
+
+---
+
+## ⚠️ STRICT: Task Matching & Skill Enforcement
+
+### Mandatory Task Classification
+
+**Before doing ANY work**, the agent MUST:
+
+1. **Classify the work** into a Task Type from the table below
+2. **Create task on task-board.md** via `task-board-management` skill ← **BLOCKING**
+3. **Load the corresponding skill** from `.github/skills/` folder
+4. **Follow the skill's execution procedure** step-by-step
+5. **Complete the skill's Definition of Done (DoD)** before marking complete
+6. **Update task-board.md** with completion status ← **MANDATORY**
+
+### Task Type Identification
+
+## Task Types Registry
+
+| Task Type | Skill | Category | Next Task | Human Review (default) |
+|-----------|-------|----------|-----------|------------------------|
+| Ideation | `task-type-ideation` | ideation-stage | Idea Mockup OR Idea to Architecture | No |
+| Idea Mockup | `task-type-idea-mockup` | ideation-stage | Requirement Gathering | No |
+| Idea to Architecture | `task-type-idea-to-architecture` | ideation-stage | Requirement Gathering | No |
+| Share Idea | `task-type-share-idea` | Standalone | - | Yes |
+| Requirement Gathering | `task-type-requirement-gathering` | requirement-stage | Feature Breakdown | Yes |
+| Feature Breakdown | `task-type-feature-breakdown` | requirement-stage | Feature Refinement | Yes |
+| Feature Refinement | `task-type-feature-refinement` | feature-stage | Technical Design | Yes |
+| Technical Design | `task-type-technical-design` | feature-stage | Test Generation | Yes |
+| Test Generation | `task-type-test-generation` | feature-stage | Code Implementation | No |
+| Code Implementation | `task-type-code-implementation` | feature-stage | Feature Closing | No |
+| Human Playground | `task-type-human-playground` | Standalone | - | Yes |
+| Feature Closing | `task-type-feature-closing` | feature-stage | User Manual | No |
+| Bug Fix | `task-type-bug-fix` | Standalone | - | Yes |
+| Refactoring Analysis | `task-type-refactoring-analysis` | code-refactoring-stage | Improve Code Quality Before Refactoring | Yes |
+| Improve Code Quality Before Refactoring | `task-type-improve-code-quality-before-refactoring` | code-refactoring-stage | Code Refactor V2 | Yes |
+| Code Refactor V2 | `task-type-code-refactor-v2` | code-refactoring-stage | - | Yes |
+| Change Request | `task-type-change-request` | Standalone | Feature Refinement OR Feature Breakdown | Yes |
+| Project Initialization | `task-type-project-init` | Standalone | Dev Environment | No |
+
+> **Note:** "Human Review (default)" is the default behavior. When **Auto-Proceed is enabled** (global or task-level), human review is **skipped** regardless of this setting.
+| Dev Environment | `task-type-dev-environment` | Standalone | - | No |
+| User Manual | `task-type-user-manual` | Standalone | - | Yes |
+
+### 🛑 STOP AND THINK: Pre-Flight Checklist
+
+**Before touching ANY code or making ANY changes, ask yourself:**
+
+```
+1. What task type is this? → Check registry table above
+2. Did I create a task on task-board.md? → If NO, STOP and create it
+3. Did I load the corresponding skill? → If NO, STOP and load it
+4. Am I following the skill's procedure? → If NO, STOP and read it
+```
+
+**Common Mistakes to Avoid:**
+- User says "refactor this" → You must use `task-type-code-refactor` skill, NOT just start coding
+- User says "fix this bug" → You must use `task-type-bug-fix` skill, NOT just fix it
+- User says "add this feature" → You must identify the right task type first
+
+### ⚠️ DO NOT Skip Skills
+
+**Forbidden Actions:**
+- ❌ Starting work without creating task on task-board.md first
+- ❌ Using `manage_todo_list` as a substitute for task-board.md
+- ❌ Completing work without updating task-board.md
+- ❌ Jumping straight to code without checking for existing tests
+- ❌ Fixing bugs without writing a failing test first
+- ❌ Implementing features without reading technical design
+- ❌ Making changes without following the skill's execution procedure
+- ❌ Refactoring code without using `task-type-code-refactor` skill
+
+**Required Actions:**
+- ✅ Always create task on task-board.md BEFORE starting work
+- ✅ Always identify task type first
+- ✅ Always load and follow the corresponding skill
+- ✅ Always check prerequisites (DoR) before starting
+- ✅ Always complete Definition of Done (DoD) before finishing
+- ✅ Always update task-board.md AFTER completing work
+
+---
+
+## Human Interaction
+
+### Approval Recognition
+
+- Do NOT assume approval unless human explicitly says: `approve`, `confirmed`, `lgtm`, `looks good`, `go ahead`, `proceed`
+- If human gives feedback without approval → keep task in current state
+- When in doubt, ask: "Do you approve this to proceed?"
+
+---
+
+## Development Principles
+
+Always follow:
+1. **SOLID** - Design principles
+3. **YAGNI** - You Aren't Gonna Need It
+4. **KISS** - Keep It Simple
+
+---
+
+## Skill Management
+
+### Creating, Updating, or Validating X-IPE Skills
+
+**When:** Creating a new skill, updating an existing skill, or validating skill structure
+**Then:** Always use the `x-ipe-skill-creator` skill
+
+```
+1. Load skill: `x-ipe-skill-creator`
+2. Follow the skill creation process defined in the skill
+3. Use appropriate template based on skill type:
+   - Task Type → templates/task-type-skill.md
+   - Skill Category → templates/skill-category-skill.md
+   - Tool Skill → TBD
+4. Validate against skill-creator checklist before completing
+```
+