Skip to content
Open
Show file tree
Hide file tree
Changes from 3 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -64,6 +64,10 @@ If you do not see the `aws-aidlc-rules` rules loaded, please check the directory

![AI-DLC Rules in Kiro CLI](./assets/images/kiro-cli-aidlc-rules-loaded.png?raw=true "AI-DLC Rules in Kiro CLI")

### Kiro Power (IDE Extension)

For Kiro IDE users, AI-DLC is also available as a Power (IDE extension) that can be installed globally or per-project. See the [Kiro Power installation guide](./aidlc-kiro-power/README.md) for detailed instructions.

### Usage

1. Start any software development project by stating your intent starting with the phrase "Using AI-DLC, ..." in the chat.
Expand Down
255 changes: 255 additions & 0 deletions aidlc-kiro-power/POWER.md
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we build this in a way that we programmatically copy over or reuse the main README.md (or parts of it) in this directory instead of duplicating its contents here. That way any changes in the main README.md will be updated in Kiro POWER.md automatically and we won't have to maintain two separate files with duplicated content. We foresee major changes in the steering files structure to address the systemic over-bloating issues, so reusing original content will be highly beneficial.

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, we can combine the install instruction into single README. I originally created a separated file to minimize the changes on the core-workflow artifacts as Kiro Power is quite specific to Kiro only.

Original file line number Diff line number Diff line change
@@ -0,0 +1,255 @@
---
name: "ai-dlc-methodology"
displayName: "AI-DLC Methodology"
description: "Comprehensive adaptive software development lifecycle that intelligently tailors workflow stages to project complexity and requirements. Guides teams through inception, construction, and operations phases with built-in quality gates and documentation."
keywords: ["ai-dlc", "development", "methodology", "lifecycle", "workflow", "adaptive", "software", "architecture"]
author: "AWS"
---

# AI-DLC Methodology

## Overview

The AI-DLC (Adaptive Intelligent Development Lifecycle) is a comprehensive software development methodology that adapts to your project's needs. Unlike rigid frameworks, AI-DLC intelligently determines which stages to execute based on:

- Project complexity and scope
- Existing codebase state (greenfield vs brownfield)
- Risk and impact assessment
- User's stated intent and clarity

The methodology consists of three main phases:
- **Inception Phase**: Planning, requirements gathering, and architectural decisions
- **Construction Phase**: Detailed design, NFR implementation, and code generation
- **Operations Phase**: Deployment and monitoring (placeholder for future expansion)

## Key Principles

- **Adaptive Execution**: Only execute stages that add value
- **Transparent Planning**: Always show execution plan before starting
- **User Control**: User can request stage inclusion/exclusion
- **Progress Tracking**: Update aidlc-state.md with executed and skipped stages
- **Complete Audit Trail**: Log ALL user inputs and AI responses in audit.md
- **Quality Focus**: Complex changes get full treatment, simple changes stay efficient
- **Content Validation**: Always validate content before file creation

## Available Steering Files

This power includes the core workflow as the main steering file:

- **core-workflow** - Main workflow orchestration and adaptive principles (use `kiroPowers` with action `readSteering`)

## Rule Detail Files

The detailed rules for each phase are located in the `steering/` folder with a flattened structure. Use `kiroPowers` tool with action `readSteering` to load these files:

### Common Rules (Always Load at Workflow Start)
- **ascii-diagram-standards.md** - ASCII diagram formatting standards
- **content-validation.md** - Content validation requirements
- **depth-levels.md** - Understanding minimal/standard/comprehensive depths
- **error-handling.md** - Error handling patterns
- **overconfidence-prevention.md** - Preventing overconfident assumptions
- **process-overview.md** - Workflow overview and adaptive principles
- **question-format-guide.md** - Question formatting rules
- **session-continuity.md** - Session resumption guidance
- **terminology.md** - Key terms and definitions
- **welcome-message.md** - Initial welcome message for users
- **workflow-changes.md** - Handling workflow modifications

### Inception Phase Rules
- **workspace-detection.md** - Detect project type and existing artifacts
- **reverse-engineering.md** - Analyze existing codebases (brownfield only)
- **requirements-analysis.md** - Gather and document requirements
- **user-stories.md** - Generate user stories and personas
- **workflow-planning.md** - Plan which stages to execute
- **application-design.md** - Design components and services
- **units-generation.md** - Break down work into units

### Construction Phase Rules
- **functional-design.md** - Design data models and business logic
- **nfr-requirements.md** - Assess non-functional requirements
- **nfr-design.md** - Design NFR patterns and solutions
- **infrastructure-design.md** - Design infrastructure and deployment
- **code-generation.md** - Generate code, tests, and artifacts
- **build-and-test.md** - Build and test instructions

### Operations Phase Rules
- **operations.md** - Deployment and monitoring (placeholder)

**Usage**:
- Use `kiroPowers` tool with action `readSteering`, powerName `ai-dlc-methodology`, steeringFile `[filename].md`
- Examples:
- `steeringFile: "core-workflow.md"` - Main workflow
- `steeringFile: "content-validation.md"` - Content validation rules
- `steeringFile: "workspace-detection.md"` - Workspace detection rules
- `steeringFile: "code-generation.md"` - Code generation rules

## How It Works

The AI-DLC methodology is defined in the `core-workflow.md` steering file, which contains:

- **Platform-specific variables** for loading rule files
- **Adaptive workflow principles** that determine which stages execute
- **Detailed stage definitions** for Inception, Construction, and Operations phases
- **Mandatory rules** for content validation, question formatting, and audit logging
- **Progress tracking** requirements and approval gates

**To understand the complete workflow**, load `core-workflow.md` using:
```
kiroPowers action: readSteering, powerName: ai-dlc-methodology, steeringFile: core-workflow.md
```

### Quick Overview

**Three-Phase Structure:**
1. **Inception Phase** - Planning, requirements, and architecture (WHAT to build)
2. **Construction Phase** - Design, implementation, and testing (HOW to build)
3. **Operations Phase** - Deployment and monitoring (placeholder for future)

**Adaptive Execution:**
- Only stages that add value are executed
- Depth levels adjust based on complexity (minimal/standard/comprehensive)
- User approval required at key gates

**Progress Tracking:**
- `aidlc-docs/aidlc-state.md` - Stage-level checkboxes
- `aidlc-docs/audit.md` - Complete interaction history

## Directory Structure

```text
<WORKSPACE-ROOT>/ # ⚠️ APPLICATION CODE HERE
├── [project-specific structure] # Varies by project
├── aidlc-docs/ # 📄 DOCUMENTATION ONLY
│ ├── inception/ # 🔵 INCEPTION PHASE
│ │ ├── plans/
│ │ ├── reverse-engineering/ # Brownfield only
│ │ ├── requirements/
│ │ ├── user-stories/
│ │ └── application-design/
│ ├── construction/ # 🟢 CONSTRUCTION PHASE
│ │ ├── plans/
│ │ ├── {unit-name}/
│ │ │ ├── functional-design/
│ │ │ ├── nfr-requirements/
│ │ │ ├── nfr-design/
│ │ │ ├── infrastructure-design/
│ │ │ └── code/ # Markdown summaries only
│ │ └── build-and-test/
│ ├── operations/ # 🟡 OPERATIONS PHASE (placeholder)
│ ├── aidlc-state.md
│ └── audit.md
```

**Critical Rule**: Application code goes in workspace root, documentation goes in aidlc-docs/

## Getting Started

1. **Start a development request** - Begin with a clear software development goal
2. **Methodology activates** - AI-DLC workflow automatically starts
3. **Follow the workflow** - Answer questions, review plans, approve stages
4. **Track progress** - Monitor `aidlc-docs/aidlc-state.md`

For detailed workflow stages and execution rules, see `core-workflow.md` steering file.

## Common Workflows

For detailed stage execution logic, see `core-workflow.md`. Here are typical patterns:

**Greenfield Project**: Workspace Detection → Requirements → User Stories → Workflow Planning → Application Design → Units Generation → Per-Unit Construction → Build and Test

**Brownfield Project**: Workspace Detection → Reverse Engineering → Requirements → Workflow Planning → Per-Unit Construction → Build and Test

**Simple Bug Fix**: Workspace Detection → Requirements (minimal) → Workflow Planning → Code Generation → Build and Test

## Best Practices

**For Users:**
- Be clear about your intent
- Review and approve plans at each gate
- Track progress in `aidlc-state.md`
- Provide feedback when requested

**For AI Agents:**
- Load `core-workflow.md` first to understand the complete workflow
- Load common rules at workflow start
- Use {RULE_LOADER_TOOL} variables for platform-specific file loading
- Follow all MANDATORY rules in core-workflow.md
- Validate content before file creation
- Log all interactions in audit.md
- Wait for explicit approval at gates

## Troubleshooting

**Workflow Not Starting**: Ensure you're making a software development request

**Missing Stages**: Stages are conditionally executed - check `workflow-planning.md` output

**Session Interrupted**: Check `aidlc-state.md` for current progress - methodology resumes from last checkpoint

**Content Validation Errors**: Review `content-validation.md` steering file for formatting rules

**Approval Gates Bypassed**: This is a bug - workflow should always wait for explicit user confirmation

For detailed troubleshooting, see relevant steering files.

## Advanced Features

Detailed in `core-workflow.md` and related steering files:

- **Multi-Package Changes** - Ordered modification sequences for brownfield projects
- **Depth Adaptation** - Automatic adjustment based on complexity and risk
- **Content Validation** - Built-in validation for Mermaid, ASCII art, and markdown
- **Question Formatting** - Standardized multiple-choice format with validation
- **Session Continuity** - Resume interrupted workflows seamlessly
- **Audit Trail** - Complete logging of all interactions

## File References

**Power Structure** (after installation):
```
ai-dlc-methodology/
├── POWER.md # Main documentation
└── steering/ # Workflow files (flattened structure)
├── core-workflow.md # Main workflow
├── ascii-diagram-standards.md # Common rules
├── content-validation.md
├── depth-levels.md
├── error-handling.md
├── overconfidence-prevention.md
├── process-overview.md
├── question-format-guide.md
├── session-continuity.md
├── terminology.md
├── welcome-message.md
├── workflow-changes.md
├── workspace-detection.md # Inception phase rules
├── reverse-engineering.md
├── requirements-analysis.md
├── user-stories.md
├── workflow-planning.md
├── application-design.md
├── units-generation.md
├── functional-design.md # Construction phase rules
├── nfr-requirements.md
├── nfr-design.md
├── infrastructure-design.md
├── code-generation.md
├── build-and-test.md
└── operations.md # Operations phase rules
```

**Generated Documentation** (created during workflow execution in your workspace):
- `aidlc-docs/aidlc-state.md` - Workflow state and progress tracking
- `aidlc-docs/audit.md` - Complete audit trail of interactions

**Source Repository**:
- Original rule files: `aidlc-rules/` directory in the repository

## Support

For issues or questions about the AI-DLC methodology:
1. Review the relevant steering file for your current stage
2. Check the troubleshooting section above
3. Review aidlc-state.md for workflow status
4. Check audit.md for interaction history
Loading