prompt-suite 1.0.13 → 1.0.15

package/README.md

@@ -20,7 +20,7 @@
 - [Technical Design Steps (Step 2)](#-technical-design-steps-step-2)
 - [Implementation Steps (Step 3)](#-implementation-steps-step-3-)
 - [Testing Steps (Steps 4-5)](#-testing-steps-steps-4-5-)
-- [Analysis Step (Step 6)](#-analysis-step-step-6)
+- [Validation Step (Step 6)](#-validation-step-step-6)
 - [Usage Tips](#-usage-tips)
   - [Input Format](#input-format)
   - [Task ID Requirement](#task-id-requirement)
@@ -118,7 +118,7 @@ If a file exists but differs from the new version, a `.new` file will be created
     ↓
 [Step 5] Test Code (Patch)
     ↓
-[Step 6] Validate & Analyze
+[Step 6] Full Validation
 ```
 
 ---
@@ -230,11 +230,11 @@ git apply prompt-suite/tasks/FEAT-001/FEAT-001-code.patch
 
 > ⚠️ **Note**: Manual `git apply` is required after Step 5.
 
-### 🔍 Analysis Step (Step 6)
+### ✅ Validation Step (Step 6)
 
-| Command         | Purpose                             | AI Behavior                          |
-| :-------------- | :---------------------------------- | :----------------------------------- |
-| `/ps.6-analyze` | Cross-stage validation (Req → Test) | **Generates** inline analysis report |
+| Command          | Purpose                             | AI Behavior                          |
+| :--------------- | :---------------------------------- | :----------------------------------- |
+| `/ps.6-validate` | Cross-stage validation (Req → Test) | **Generates** inline analysis report |
 
 ---
 
@@ -445,7 +445,7 @@ git apply prompt-suite/tasks/FEAT-001/FEAT-001-test.patch
 #### Step 6: Validate
 
 ```bash
-/ps.6-analyze \
+/ps.6-validate \
   prompt-suite/tasks/FEAT-001/FEAT-001-requirement.md \
   prompt-suite/tasks/FEAT-001/FEAT-001-design.md \
   prompt-suite/tasks/FEAT-001/FEAT-001-test.md

package/docs/AI-Driven Software Development Workflow.md

@@ -0,0 +1,209 @@
+# AI-Driven Software Development Workflow
+
+_(Customized for real-world SDLC)_
+
+## 1. Mastering the SDLC: The AI Leap
+
+**1. Rules** | **2. Requirements** | **3. Design** | **4. Code** | **5. Test Plan** | **6. Test Code** | **7. Validation**
+
+### The Paradigm Shift
+
+- **The Manual Legacy**: Historically, these stages were siloed and executed manually, often resulting in fragmented documentation and structural inconsistencies.
+- **The AI-Optimized Future**: We are redefining the lifecycle by embedding AI into every phase. Through **"Standardized Prompts"**, we transform the manual grind into a high-velocity, high-quality pipeline.
+
+---
+
+## 2. Applying AI to the SDLC
+
+### AI Components
+
+- **AI Models**
+  Gemini / Claude / GPT / Grok …
+- **AI Agents**
+  Context-aware tools, custom agents… (context7, Serena...)
+- **Prompt Engineering**
+  **Precise – Context-rich – Output-constrained** prompts.
+
+> AI is only powerful when **input is structured** and **output is controlled**.
+
+---
+
+## 3. Core Strategy
+
+> **Apply the Spec-kit mindset**, but **re-design prompts according to the SDLC**.
+>
+> Each step in the process = **1 clear command, 1 standardized output**.
+
+This integrated system is called **Prompt Suite (ps)**.
+
+---
+
+## 4. Prompt Suite – Detailed Workflow
+
+### Step 0 – Coding Rules (One-time Setup)
+
+Set the "rules of the game" for the entire project.
+
+| Command           | Purpose                           |
+| ----------------- | --------------------------------- |
+| `/ps.0-gen-rules` | Generate coding rules per project |
+| `/ps.0-rev-rules` | Review & refine rules             |
+
+**Output**
+
+```text
+prompt-suite/
+ └── rules/
+     ├── architecture.md
+     ├── backend.md
+     ├── frontend.md
+     ├── database.md
+     └── ...
+```
+
+> Perform this **once**, reuse it throughout the project lifecycle.
+
+---
+
+### Step 1 – Requirements
+
+| Command                 | Purpose                              |
+| ----------------------- | ------------------------------------ |
+| `/ps.1-gen-requirement` | Raw request → structured requirement |
+| `/ps.1-rev-requirement` | Check for gaps and inconsistencies   |
+
+**Output**
+`{task-id}-requirement.md`
+
+---
+
+### Step 2 – Technical Design
+
+| Command            | Purpose                        |
+| ------------------ | ------------------------------ |
+| `/ps.2-gen-design` | Requirement → technical design |
+| `/ps.2-rev-design` | Review technical feasibility   |
+
+**Output**
+`{task-id}-design.md`
+
+---
+
+### Step 3 – Feature Code
+
+| Command          | Purpose                   |
+| ---------------- | ------------------------- |
+| `/ps.3-gen-code` | Design → code (git patch) |
+| `/ps.3-rev-code` | Review code quality       |
+
+**Output**
+`{task-id}-feature.patch`
+
+⚠️ **Manual `git apply` required**
+→ keeps the developer in full control.
+
+---
+
+### Step 4 – Test Plan
+
+| Command               | Purpose                 |
+| --------------------- | ----------------------- |
+| `/ps.4-gen-test-plan` | Design → test scenarios |
+| `/ps.4-rev-test-plan` | Check test coverage     |
+
+**Output**
+`{task-id}-test.md`
+
+---
+
+### Step 5 – Test Code
+
+| Command               | Purpose               |
+| --------------------- | --------------------- |
+| `/ps.5-gen-test-code` | Test plan → test code |
+| `/ps.5-rev-test-code` | Review test quality   |
+
+**Output**
+`{task-id}-test.patch`
+
+⚠️ **Manual `git apply` required**
+
+---
+
+### Step 6 – Validation
+
+| Command          | Purpose                  |
+| ---------------- | ------------------------ |
+| `/ps.6-validate` | Validate the entire flow |
+
+**Output**
+
+- Inconsistencies
+- Missing cases
+- Design–code mismatch
+- Risks & improvement suggestions
+
+---
+
+## 5. End-to-End Flow Diagram
+
+```text
+
+       ┌──────────────────────┐        ┌──────────────────────┐        ┌──────────────────────┐
+       │   STEP 0: RULES      │        │  STEP 1: REQUIREMENT │        │   STEP 2: DESIGN     │
+       ├──────────────────────┤        ├──────────────────────┤        ├──────────────────────┤
+       │ AI: Detect & Gen     │ ──────>│ AI: Structure Req    │ ──────>│ AI: Tech Blueprint   │
+       │ OUT: /rules/         │        │ OUT: requirement.md  │        │ OUT: design.md       │
+       │ DEV: Review Rules    │        │ DEV: Approve Logic   │        │ DEV: Approve Arch    │
+       └──────────────────────┘        └──────────────────────┘        └──────────┬───────────┘
+                                                                                  │
+             ┌────────────────────────────────────────────────────────────────────┘
+             │
+             ▼
+       ┌──────────────────────┐        ┌──────────────────────┐        ┌──────────────────────┐
+       │   STEP 3: CODE       │        │  STEP 4 & 5: TEST    │        │   STEP 6: VALIDATION │
+       ├──────────────────────┤        ├──────────────────────┤        ├──────────────────────┤
+       │ AI: Gen Patch        │ ──────>│ AI: Gen Test Patch   │ ──────>│ AI: Full Validation  │
+       │ OUT: feature.patch   │        │ OUT: test.patch      │        │ OUT: Final Report    │
+       │ DEV: Review & Apply  │        │ DEV: Review & Apply  │        │ DEV: Task Sign-off   │
+       └──────────────────────┘        └──────────────────────┘        └──────────────────────┘
+```
+
+---
+
+## 6. Flexible Workflow – Pipeline Options
+
+### 🔹 Small task (bug fix)
+
+`Requirement → Design → Code`
+
+Commands:
+`/ps.1-gen-requirement`
+`/ps.2-gen-design`
+`/ps.3-gen-code`
+
+---
+
+### 🔹 Medium task (new feature)
+
+`Requirement → Review → Design → Review → Code → Test Code`
+
+---
+
+### 🔹 Large / Critical feature
+
+`Full workflow (Steps 0 → 6)`
+
+---
+
+## 7. Value Proposition
+
+- Standardizes technical development mindset.
+- AI follows structured **artifacts**, not random coding.
+- End-to-end traceability:
+  **Requirement → Design → Code → Test**
+- Ideal for:
+  - Team scaling
+  - Developer onboarding
+  - Review & Auditing
+  - Controlled AI-assisted development.

package/package.json

@@ -1,20 +1,11 @@
 {
   "name": "prompt-suite",
-  "version": "1.0.13",
+  "version": "1.0.15",
   "description": "Prompt Suite - Prompt Engineering & Workflow Automation",
   "main": "index.js",
   "bin": {
     "prompt-suite": "bin/cli.js"
   },
-  "files": [
-    "bin/",
-    "prompts/",
-    "rules/",
-    "scripts/",
-    "tasks/",
-    "index.js",
-    "README.md"
-  ],
   "scripts": {
     "test": "echo \"Error: no test specified\" && exit 1"
   },

package/prompts/ps.5-rev-test-code.prompt.md

@@ -168,4 +168,4 @@ STRICT OUTPUT RULE:
 
 ## Next Steps
 
-After reviewing test code → run `/ps.6-analyze` to validate entire flow
+After reviewing test code → run `/ps.6-validate` to validate entire flow

package/prompts/{ps.6-analyze.prompt.md → ps.6-validate.prompt.md}

@@ -1,8 +1,8 @@
 ---
-description: Validate requirement → design → code → test flow, find inconsistencies
+description: Full flow validation (Requirement → Design → Code → Test)
 ---
 
-# /ps.6-analyze
+# /ps.6-validate
 
 ## Description
 
@@ -14,8 +14,8 @@ Validate the entire flow from requirement → tech design → code → test, det
 
 - Provide paths to your requirement, design, and test plan files
 - **Task-id extraction**: AI will try to extract from filenames (optional)
-  - If found, will be used in analysis report
-  - If not found, analysis will proceed without task-id reference
+  - If found, will be used in validation report
+  - If not found, validation will proceed without task-id reference
 - **Files can be located anywhere** (not limited to `/prompt-suite`)
 - Examples:
   - `~/FEAT-001-req.md ~/FEAT-001-design.md ~/FEAT-001-test.md` ✅ (with task-id)
@@ -135,7 +135,7 @@ SELF-CHECK BEFORE OUTPUT:
 5. ✅ Report follows the structured format.
 
 STRICT OUTPUT RULE:
-- Output ONLY the Analysis Report.
+- Output ONLY the Validation Report.
 - No meta commentary before or after the report.
 - No summaries or introductions.
 ```
@@ -148,7 +148,7 @@ STRICT OUTPUT RULE:
 4. ✅ **Coverage** - Check if coverage is sufficient
 5. ✅ **Prioritize** - P0/P1/P2 recommendations
 
-## Analysis Checklist
+## Validation Checklist
 
 ### Requirement → Tech Design
 
@@ -189,19 +189,19 @@ STRICT OUTPUT RULE:
 
 ```bash
 # 1. Run with file paths (can be anywhere in filesystem)
-/ps.6-analyze \
+/ps.6-validate \
   /path/to/FEAT-001-requirement.md \
   /path/to/FEAT-001-design.md \
   /path/to/FEAT-001-test.md
 
-# 2. AI reads all files and analyzes
-# 3. Receive comprehensive analysis report
+# 2. AI reads all files and validates
+# 3. Receive comprehensive validation report
 ```
 
 ## Example Output Structure
 
 ```markdown
-# Analysis Report: FEAT-001
+# Validation Report: FEAT-001
 
 ## 1. Alignment Check
 
@@ -276,4 +276,4 @@ STRICT OUTPUT RULE:
 
 ## Next Steps
 
-After analyze → Fix issues by priority → Ready to deploy!
+After validation → Fix issues by priority → Ready to deploy!