agentic-sdlc 1.0.0

package/docs/guides/MCP-GUIDE.md

@@ -0,0 +1,53 @@
+# 🌐 Distributed Intelligence Network: MCP Integration Guide
+
+This project leverages the **Model Context Protocol (MCP)** to provide agents with real-world tools, system access, and external data.
+
+## 🛠️ Core MCP Servers
+
+The following servers are integrated into the team roles. Ensure these are configured in your environment for full automation.
+
+| MCP Server | Key Tools | Primary Role |
+| :--- | :--- | :--- |
+| **GitHub** | Issue tracking, Milestones, PRs | @PM, @PO, @REPORTER |
+| **Vercel** | Deployment status, Logs | @DEVOPS |
+| **Docker** | Container management | @DEVOPS |
+| **GitIngest** | Codebase snapshots | @ORCHESTRATOR, @REPORTER |
+| **Apidog** | API Testing & Design | @SA, @TESTER |
+| **Brave Search** | External Research | @PM, @PO |
+| **Firecrawl** | Web Scraper / Log research | @SECA, @DEVOPS |
+| **Playwright** | E2E / Browser Testing | @QA, @TESTER |
+| **Context7** | Architecture Analysis | @SA, @DEV |
+| **Supabase / PG** | Database Operations | @SA, @DEV |
+| **Sequential Thinking** | Logic & Reasoning | @ORCHESTRATOR, @DEV |
+| **DesktopCommander** | OS Management | @ORCHESTRATOR, @DEVOPS |
+
+---
+
+## 🚀 Advanced MCP Suggestions (Future-Proofing)
+
+Beyond the requested list, the following MCPs are highly recommended for advanced project scaling:
+
+### 1. Productivity & Handoffs
+- **Slack/Discord MCP:** Push real-time role handoff notifications (e.g., `@QA: Design is ready for review`).
+- **Linear MCP:** If GitHub Issues become too complex, Linear offers a high-performance alternative for engineering tasks.
+- **Sentry MCP:** Automatically pull crash reports into @DEV's context for instant bug analysis.
+
+### 2. Knowledge & AI Enhancement
+- **Exa / Perplexity MCP:** For "AI-Search" that provides factual, cited answers faster than a standard search engine.
+- **Zotero / Mendeley MCP:** If the project requires heavy academic or technical research documentation.
+- **Obsidian / Logseq MCP:** Personal knowledge management sync for stakeholders.
+
+### 3. Data & Cloud
+- **Snowflake / Databricks MCP:** For heavy data engineering projects.
+- **AWS / GCP / Azure MCPs:** To manage infrastructure directly from the chat.
+
+---
+
+## 📋 Best Practices for Agents
+
+1. **Facts First:** Before starting a task, use `Brave Search` or `GitIngest` to verify the current state.
+2. **Atomic Integration:** When mentioning an MCP tool, provide the context (e.g., *"According to Playwright logs, the login button is obscured..."*).
+3. **Loop Verification:** Always use `Apidog` or `Playwright` to verify a fix before tagging the next role.
+
+---
+*Created by Antigravity - 2026-01-01*

package/docs/guides/QUICK-START.md

@@ -0,0 +1,104 @@
+# Quick Start Guide
+
+Get started with `agentic-sdlc` in 5 minutes.
+
+---
+
+## 🚀 Installation
+
+```bash
+npm install -g agentic-sdlc
+```
+
+---
+
+## ⚡ 3 Steps to Start
+
+### Step 1: Create Project (30 seconds)
+
+```bash
+agentic-sdlc create my-project
+cd my-project
+```
+
+### Step 2: Setup IDE (15 seconds)
+
+```bash
+agentic-sdlc ide cursor
+```
+
+### Step 3: Start Building (Instant)
+
+Open your IDE and type:
+```
+/pm Build a todo app with authentication
+```
+
+**That's it!** 🎉
+
+---
+
+## 🎯 What You Get
+
+After setup, you have:
+
+✅ **12 AI Roles** ready to use:
+- `/pm` - Project Manager
+- `/sa` - System Analyst  
+- `/uiux` - UI/UX Designer
+- `/dev` - Developer
+- `/tester` - Tester
+- And 7 more...
+
+✅ **16 Templates** for documentation
+
+✅ **Knowledge Base** for learning
+
+✅ **IDE Integration** with slash commands
+
+---
+
+## 💡 Try These Commands
+
+```bash
+# Full automation (fastest)
+/auto Create a mobile fitness app
+
+# Manual control (learn the flow)
+/pm Build a blog platform
+
+# Specific tasks
+/dev Implement OAuth2 authentication
+/uiux Design mobile dashboard
+/devops Setup CI/CD pipeline
+
+# Search knowledge
+/kb-search React hooks error
+```
+
+---
+
+## 📚 Learn More
+
+- **Full Examples:** See `CLI-EXAMPLES.md`
+- **Usage Guide:** `.agent/usage.md`
+- **All Commands:** `agentic-sdlc --help`
+
+---
+
+## 🆘 Need Help?
+
+```bash
+# Show all commands
+agentic-sdlc --help
+
+# List available resources
+agentic-sdlc list
+
+# Check version
+agentic-sdlc --version
+```
+
+---
+
+**Ready to build? Let's go! 🚀**

package/docs/reports/comparison-leann-neo4j.md

@@ -0,0 +1,49 @@
+# Comparative Analysis: LEANN vs. Neo4j (AI Brain Architecture)
+**Author:** @SA (System Analyst) | **Project:** Template Instructions | **Date:** 2026-01-01
+
+## 📊 Summary Overview
+While both technologies serve as an AI "memory," they operate on fundamentally different principles. **LEANN** is optimized for high-performance semantic search (Vector RAG), while **Neo4j** is built for complex relationship discovery (Graph RAG).
+
+| Feature | **LEANN** (Vector Memory) | **Neo4j** (Graph Brain) |
+|:---|:---|:---|
+| **Primary Goal** | Finding *similar* text/code. | Finding *connections* between entities. |
+| **Logic** | Semantic Similarity (Cosine Similarity). | Pathfinding & Relational Logic (Cypher). |
+| **Data Structure** | Flattened Embedding Vectors. | Nodes (Entities) and Edges (Relationships). |
+| **Best For** | "Find a code snippet like this." | "Why did changing $X$ cause a bug in $Y$?" |
+| **Storage** | Ultra-lightweight (97% savings). | Moderate (requires structured DB). |
+| **Performance** | Extremely fast retrieval. | Fast for deep path traversal. |
+| **Platform** | Local (Python/CLI/WSL2). | Server/Docker based. |
+
+---
+
+## 🔍 Deep Dive Comparison
+
+### 1. LEANN: The "Librarian"
+LEANN acts like a highly efficient librarian. It indexes the "content" of your files using AST-aware chunking.
+*   **Strengths:** It is the best at answering "Where is the logic for X?" It excels at retrieving large amounts of context for an AI prompt with minimal resource overhead.
+*   **Weaknesses:** It has no concept of "logic flow" or "historical cause-and-effect." It treats every chunk of code as an isolated island.
+
+### 2. Neo4j: The "Detective"
+Neo4j acts like a detective with a "evidence board" (red strings connecting photos).
+*   **Strengths:** It can map the entire project's lineage. It understands that `Function A` belongs to `File B`, was written by `Role:Dev` during `Sprint 2`, and fixed `Issue #42`. This is the foundation of **Self-Learning AI**, as it allows the AI to traverse paths of success and failure.
+*   **Weaknesses:** It is not a search engine. Asking Neo4j "find code that looks like this" is inefficient without a supporting vector index (like LEANN).
+
+---
+
+## 💡 The "Hybrid" Recommendation
+For a truly "Self-Improving" project, you should not choose one over the other. Instead, use them in **Tandem**:
+
+1.  **LEANN** handles the **Retrieval** (finding the raw code/text).
+2.  **Neo4j** handles the **Reasoning** (explaining the relationships between those snippets).
+
+### Suggested Architecture
+*   **Step 1:** LEANN finds 3 relevant code snippets.
+*   **Step 2:** The AI queries Neo4j: *"What is the history of these snippets? What bugs are they linked to?"*
+*   **Step 3:** The AI combines the code + history to provide a perfect, "learned" solution.
+
+---
+### Next Steps
+1. **Approve** the Neo4j Sprint Plan to begin the Docker setup.
+2. We will implement a bridge script that uses LEANN's output to "populate" the Neo4j graph.
+
+#analysis #leann #neo4j #architecture #ai-memory

package/docs/setup/github-management.md

@@ -0,0 +1,37 @@
+# GitHub Labels for AI SDLC Management
+
+To effectively manage this project using the AI roles, use the following labels:
+
+## 🏷️ Type Labels
+- `type:bug`: Something is broken.
+- `type:feature`: Request for new functionality.
+- `type:task`: Technical implementation work.
+- `type:security`: Vulnerability report.
+- `type:docs`: Documentation updates.
+
+## 👤 Role Labels
+- `role:pm`: Project Management.
+- `role:po`: Product Owner.
+- `role:sa`: System Analyst.
+- `role:uiux`: UI/UX Designer.
+- `role:dev`: Developer.
+- `role:qa`: Quality Assurance.
+- `role:tester`: Testing.
+- `role:devops`: Operations / CI/CD.
+- `role:seca`: Security Analyst.
+
+## 🚦 Priority Labels
+- `priority:p0`: Critical / Blocker.
+- `priority:p1`: High Priority.
+- `priority:p2`: Medium Priority.
+- `priority:p3`: Low Priority.
+
+## 🔄 Status Labels
+- `status:triage`: New issue needs review.
+- `status:backlog`: Approved but not started.
+- `status:in-progress`: Currently being worked on.
+- `status:in-review`: Waiting for @QA or @Stakeholder.
+- `status:done`: Completed.
+
+---
+*Generated for the template-instructions manager.*

package/docs/sprints/sprint-github-issues.md

@@ -0,0 +1,36 @@
+# Implementation Plan: GitHub Issue Management Configuration
+
+This plan outlines the setup of GitHub Issue templates and organization to align with the AI-driven SDLC roles in the `template-instructions` project.
+
+## 🎯 Objectives
+- Standardize issue reporting for different SDLC roles.
+- Enable seamless hand-off between AI agents using issue numbers.
+- Provide a structured backlog for the `@PM` and `@PO` to manage.
+
+## 🛠️ Proposed Changes
+
+### 1. GitHub Issue Templates
+Create `.github/ISSUE_TEMPLATE/` folder with the following:
+- `bug_report.yml`: For `@Tester` and `@QA` to report issues.
+- `feature_request.yml`: For `@PO` and `@Stakeholder` to propose ideas.
+- `task_implementation.yml`: For `@PM` to assign work to `@Dev`.
+- `security_alert.yml`: For `@SECA` to report vulnerabilities.
+- `config.yml`: Global configuration for templates.
+
+### 2. Labeling System
+Define a set of labels that match the project's specialized roles:
+- `role:pm`, `role:dev`, `role:qa`, `role:sa`, etc.
+- `type:bug`, `type:feature`, `type:task`, `type:security`.
+- `priority:p0`, `priority:p1`, `priority:p2`.
+
+### 3. Role Integration
+Update `@Reporter` and `@PM` workflow instructions to mention creating/updating GitHub issues as part of their status reports.
+
+## 📅 Step-by-Step Execution
+
+1. Create the `.github/ISSUE_TEMPLATE` directory.
+2. Generate the YAML-based issue templates (more modern than MD).
+3. Update the Project Management documentation to include "GitHub Issues" as the source of truth for the backlog.
+
+---
+**Status:** ⏳ Pending approval from USER.

package/docs/sprints/sprint-leann-integration.md

@@ -0,0 +1,41 @@
+# Implementation Plan: LEANN AI Brain Integration
+
+This plan outlines the steps to integrate the **LEANN** (Lean and Efficient AI Near-memory) framework into the `template-instructions` project to automate knowledge management and provide agents with deep project context.
+
+## 🎯 Objectives
+- Replace manual Knowledge Base indexing with automated LEANN indexing.
+- Enable deep, AST-aware code search for AI roles.
+- Integrate LEANN via MCP for real-time memory access in the IDE.
+
+## 🛠️ Proposed Changes
+
+### 1. New Workflow: `leann-brain.md`
+Create a new workflow in `.agent/workflows/` to handle installation, indexing, and querying.
+
+### 2. Update `kb-search.md`
+Modified to prefer LEANN search over manual file grepping when available.
+
+### 3. Documentation
+Update `README.md` and create `docs/brain.md` to explain how to use the new "Project Brain".
+
+### 4. Role Updates
+Update the AI role instructions (PM, Dev, SA) to include the `/brain` command as their primary context source.
+
+## 📅 Step-by-Step Execution
+
+### Phase 1: Setup
+1. **Command:** `pip install leann-core leann-backend-hnsw`
+   - *Note: Suggest using `uv` if available for speed.*
+2. **Command:** `leann index --path .`
+   - Performs initial AST-aware indexing of the workspace.
+
+### Phase 2: Workflow Creation
+1. Create `.agent/workflows/leann-brain.md`.
+2. Update `.agent/workflows/kb-search.md`.
+
+### Phase 3: Integration
+1. Configure MCP server for the IDE.
+2. Update global rules to encourage using `/brain` for context.
+
+---
+**Status:** ⏳ Pending approval from USER.

package/docs/sprints/sprint-neo4j-brain.md

@@ -0,0 +1,38 @@
+# Project Plan: Neo4j AI Self-Learning Brain
+**Sprint:** 1 | **Version:** 1.0 | **Status:** ⏳ Strategy Phase
+
+## 🎯 Vision
+Transform the static project knowledge into a **Dynamic Graph Brain** using Neo4j. This will allow the AI agents to not only "search" for text but to "understand" relationships between concepts, code modules, and past bugs to automatically improve their coding skills.
+
+## 🛠️ Tech Stack
+- **Graph Database:** Neo4j AuraDB (Cloud-based / Managed Service)
+- **Integration Layer:** Python + LangChain (Graph-QA)
+- **Knowledge Source:** Local workspace + Git history + LEANN Index
+- **Brain Logic:** Cypher Query Language (CQL) for relationship discovery
+
+## 🚀 Key Features
+1. **Relationship Mapping:** Link `Issue #123` -> `Modified Function X` -> `Introduced Bug Y`.
+2. **Skill Progression Graph:** Track which AI roles are most effective in specific modules.
+3. **Automated "Self-Improvement" Loop:** 
+   - AI analyzes failing tests in Neo4j.
+   - Finds the "Reasoning Path" from previous similar fixes.
+   - Suggests optimized implementation strategies.
+
+## 📅 Roadmap (Must-Have)
+- [ ] **Phase 1: Cloud Provisioning** (Create Free Instance on Neo4j Aura)
+- [ ] **Phase 2: Data Modeling** (Define Nodes: `Code`, `Role`, `Issue`, `Concept`)
+- [ ] **Phase 3: Secure Connection** (Configure `.env` with Aura credentials)
+- [ ] **Phase 4: Ingestion Pipeline** (Export LEANN and Git data to Aura)
+- [ ] **Phase 5: Brain Query Interface** (Add `/graph-query` command)
+
+## ⚠️ Risks
+- **Complexity:** Graph modeling requires careful schema design.
+- **Resource Usage:** Neo4j requires more RAM than simple vector search.
+
+---
+### Next Step:
+- **USER:** Please create a free instance at [Neo4j Aura](https://neo4j.com/cloud/aura/) and provide the **URI**, **Username**, and **Password**.
+- **@SA:** Standing by to design the Neo4j Schema for Cloud ingestion.
+- **@DevOps:** Standing by to setup the Python connection client.
+
+#planning #self-learning #neo4j #ai-brain

package/package.json

@@ -0,0 +1,16 @@
+{
+  "name": "agentic-sdlc",
+  "version": "1.0.0",
+  "description": "Simulating a complete Software Development Lifecycle (SDLC) with specialized AI Agents.",
+  "main": "index.js",
+  "bin": {
+    "agentic-sdlc": "./bin/cli.js"
+  },
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "dependencies": {
+    "fs-extra": "^11.2.0"
+  },
+  "type": "module"
+}