@iservu-inc/adf-cli 0.2.0 → 0.3.6

package/CHANGELOG.md

@@ -7,6 +7,331 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.3.6] - 2025-10-04
+
+### ⚙️ New Configuration Command
+
+**New `adf config` command:**
+- Centralized configuration management with category-based menu
+- **AI Provider Setup** as first configuration category
+- Green ✓ status indicators for configured items
+- Yellow ○ indicators for unconfigured items
+- Shows current provider name when already configured
+- Easy reconfiguration - run anytime to switch providers or models
+- Extensible design for future configuration categories
+
+**AI Configuration Made Optional:**
+- AI configuration now optional during `adf init`
+- Prompt: "Configure AI provider now? (Enables intelligent follow-up questions)"
+- Can skip and configure later with `adf config`
+- Helpful reminder: "💡 You can configure AI later by running: adf config"
+- Interview works with or without AI (graceful degradation)
+
+### 🎯 UX Improvements
+
+**User Experience Enhancements:**
+- Added visible "skip" instruction hint to every question
+  - Displays `(Type "skip" to skip remaining questions in this block)` below each question
+  - Prevents user frustration by making escape route obvious
+  - Users no longer need to guess how to skip questions
+
+- Improved deployment messaging for clarity
+  - Changed confusing "Next Steps" to clear "Requirements Complete!"
+  - New prompt: "Automatically deploy to your IDE? (I'll configure everything for you)"
+  - Default changed to Yes to encourage automation
+  - Removed misleading manual steps that made deployment seem manual
+  - Emphasizes that adf-cli does ALL the work automatically
+
+**Impact:**
+- More flexible configuration workflow
+- Better user guidance throughout interview process
+- Clearer expectations about automated deployment
+- Reduced confusion and frustration
+
+## [0.3.5] - 2025-10-04
+
+### 🔧 Enhanced AI Configuration
+
+**API Key Persistence (.env):**
+- API keys now saved to `.adf/.env` file (project-local, gitignored)
+- Automatic loading on subsequent runs
+- No need to set environment variables manually
+- Multiple provider keys supported in same file
+- Secure storage with clear warnings
+
+**Dynamic Model Fetching:**
+- **OpenAI**: Fetches real models from your account API
+- **OpenRouter**: Fetches 100+ available models dynamically
+- **Anthropic/Google**: Uses curated default models
+- Graceful fallback to defaults if API call fails
+
+**Autocomplete Model Selection:**
+- Type to filter models in real-time (e.g., type "gpt-4" to see only GPT-4 models)
+- Arrow keys navigate filtered results
+- Shows 10 models at a time for easy browsing
+- Enter to select highlighted model
+
+**New Dependencies:**
+- `inquirer-autocomplete-prompt@2.0.1` - Autocomplete functionality
+- `dotenv@17.2.3` - .env file management
+- `node-fetch@2.7.0` - API calls for model fetching
+
+**User Experience:**
+```
+First time:
+? Enter your API key: [hidden]
+✓ API key saved to: .adf/.env
+⠹ Fetching available models...
+? Select model (type to filter): |
+
+Next time (same project):
+✓ Detected API keys for: Anthropic Claude
+✓ Using API key from .env file
+```
+
+## [0.3.4] - 2025-10-04
+
+### 🤖 Multi-Provider AI Integration
+
+**Major Enhancement: 4 AI Provider Support**
+
+This is a **paradigm shift** - AI connectivity is now **REQUIRED** before the interview starts, enabling real-time answer analysis and intelligent follow-ups.
+
+**Supported Providers:**
+- **Anthropic Claude** (Sonnet 4.5, Claude 3.5, Opus)
+- **OpenAI GPT** (GPT-4-turbo, GPT-4o, GPT-4, GPT-3.5-turbo)
+- **Google Gemini** (2.0-flash-exp, 1.5-pro, 1.5-flash)
+- **OpenRouter** (Multi-model access to 100+ models)
+
+**AI-Powered Features:**
+
+1. **Real-Time Answer Quality Analysis**
+   - AI evaluates every answer for specificity, completeness, clarity, technical depth
+   - Provides 0-100 quality score instantly
+   - Shows suggestions for improvement
+   - Identifies missing elements
+
+2. **Intelligent Follow-Up Questions**
+   - AI generates contextual follow-ups based on answer quality
+   - Questions are specific to what's missing, not generic
+   - Adapts to user's knowledge level and project context
+   - Fallback to heuristic follow-ups if AI unavailable
+
+3. **Provider Configuration Wizard**
+   - Interactive provider selection
+   - API key validation with format checking
+   - Model selection per provider
+   - Connection testing before interview starts
+   - Secure API key handling (never committed)
+
+**New Files:**
+- `lib/ai/ai-config.js` - Provider configuration and setup
+- `lib/ai/ai-client.js` - Unified AI client for all providers
+
+**New Dependencies:**
+- `@anthropic-ai/sdk@0.32.0` - Anthropic Claude support
+- `@google/generative-ai@0.21.0` - Google Gemini support
+- `openai@4.73.0` - OpenAI GPT support
+
+**Interview Flow Changes:**
+```
+OLD:
+1. Detect project type
+2. Select framework
+3. Start interview
+
+NEW:
+1. Detect project type
+2. Select framework
+3. Configure AI Provider (REQUIRED)
+   - Select provider
+   - Enter API key
+   - Test connection
+4. Start AI-guided interview with real-time analysis
+```
+
+**Graceful Degradation:**
+- If AI analysis fails, automatically falls back to heuristic scoring
+- Interview never blocked by AI issues
+- Clear error messages with troubleshooting guidance
+
+## [0.3.0] - 2025-10-03
+
+### 🚀 MAJOR RELEASE: Tool Configuration Generators
+
+This release adds comprehensive tool configuration generation for all major AI coding environments. The ADF CLI now automatically generates optimized configurations for Windsurf, Cursor, VS Code, and the universal AGENTS.md standard.
+
+### Added
+
+**Universal Standard:**
+- **AGENTS.md Generator** - Universal AI agent configuration standard
+  - Works with OpenAI Codex, Cursor, Windsurf, Google Gemini/Jules, Factory
+  - Supported by 20,000+ open-source projects
+  - Automatically generated from framework outputs (PRP, Balanced, BMAD)
+  - Nested file support for monorepos
+
+**Windsurf (Codeium) Generator:**
+- `.windsurfrules` - Legacy format with 12,000 character limit compliance
+- `.windsurf/rules/` - Modern modular rules system
+  - `project-context.md` - Project overview and tech stack
+  - `architecture.md` - System architecture patterns
+  - `coding-standards.md` - Code style and best practices
+- `.windsurf/workflows/` - Cascade AI workflows
+  - `review-requirements.md` - Requirements review workflow
+  - `implement-feature.md` - Feature implementation workflow (Balanced/BMAD)
+
+**Cursor AI Generator:**
+- `.cursor/rules` - Modern primary configuration format
+- `.cursorrules` - Legacy deprecation notice with migration path
+- `.cursor/mcp.json` - Model Context Protocol support (optional, prepared for future)
+- Custom modes: architect, implementer, reviewer
+- Framework-specific context and instructions
+
+**VS Code Generator:**
+- `.github/copilot-instructions.md` - GitHub Copilot instructions
+- `.vscode/settings.json` - Custom chat modes
+  - Architect mode - Focus on system design
+  - Implementer mode - Feature implementation with TDD
+  - Reviewer mode - Code review against requirements
+- Preserves existing settings while adding chat modes
+- Language Model API integration ready
+
+**Base Architecture:**
+- `ToolConfigGenerator` - Base class with shared utilities
+  - Framework output loading (PRP, Balanced, BMAD)
+  - Section extraction and parsing
+  - Template variable substitution
+  - Project metadata detection
+
+**Integration:**
+- Enhanced `adf deploy` command to generate all configurations
+  - `adf deploy windsurf` - Generates Windsurf configs + AGENTS.md
+  - `adf deploy cursor` - Generates Cursor configs + AGENTS.md
+  - `adf deploy vscode` - Generates VS Code configs + AGENTS.md
+  - AGENTS.md always generated (universal standard)
+- Automatic session detection and framework identification
+- Error recovery with warning messages
+
+**Comprehensive Testing:**
+- 24 new unit tests for generators (57 total tests)
+- Test coverage maintained at 78%
+- Tests for all 3 frameworks (PRP, Balanced, BMAD)
+- Tests for all 3 tool generators
+- Template variable substitution tests
+- File preservation tests (VS Code settings)
+
+### New Files
+
+**Generators:**
+- `lib/generators/tool-config-generator.js` - Base class for all generators
+- `lib/generators/agents-md-generator.js` - Universal AGENTS.md standard
+- `lib/generators/windsurf-generator.js` - Windsurf configuration generator
+- `lib/generators/cursor-generator.js` - Cursor configuration generator
+- `lib/generators/vscode-generator.js` - VS Code configuration generator
+- `lib/generators/index.js` - Generator exports and unified API
+
+**Tests:**
+- `tests/agents-md-generator.test.js` - 5 tests for AGENTS.md
+- `tests/windsurf-generator.test.js` - 7 tests for Windsurf
+- `tests/cursor-generator.test.js` - 6 tests for Cursor
+- `tests/vscode-generator.test.js` - 6 tests for VS Code
+
+**Documentation:**
+- `.project/docs/tool-integrations/RESEARCH-FINDINGS.md` - 500+ line research document
+- `.project/docs/tool-integrations/IDE-CUSTOMIZATIONS.md` - Tool customization guide
+- `.project/chats/current/2025-10-03_AGENTS-MD-AND-TOOL-GENERATORS.md` - Session documentation
+
+### Changed
+
+**Deploy Command:**
+- Enhanced to call tool-specific generators based on selected tool
+- Added progress messages for each generation phase
+- Improved error handling with graceful fallbacks
+- Displays generated file counts and paths
+
+**Framework Support:**
+- All generators support PRP (Rapid) framework
+- All generators support Balanced framework
+- All generators support BMAD (Comprehensive) framework
+- Automatic framework detection from session metadata
+
+### Generated File Structure
+
+When running `adf deploy <tool>`, the following files are created:
+
+**Always Generated:**
+```
+AGENTS.md                                    # Universal standard
+```
+
+**Windsurf:**
+```
+.windsurfrules                               # Legacy format
+.windsurf/
+├── rules/
+│   ├── project-context.md
+│   ├── architecture.md
+│   └── coding-standards.md
+└── workflows/
+    ├── review-requirements.md
+    └── implement-feature.md                 # Balanced/BMAD only
+```
+
+**Cursor:**
+```
+.cursor/
+├── rules                                    # Primary config
+└── mcp.json                                 # Optional (prepared)
+.cursorrules                                 # Deprecation notice
+```
+
+**VS Code:**
+```
+.github/
+└── copilot-instructions.md                  # GitHub Copilot
+.vscode/
+└── settings.json                            # Custom chat modes
+```
+
+### Technical Details
+
+**Template Variables:**
+- `PROJECT_NAME` - Project name from metadata
+- `SESSION_ID` - Unique session identifier
+- `FRAMEWORK` - rapid/balanced/comprehensive
+- Framework-specific variables (PRP_GOAL, CONSTITUTION_PRINCIPLES, PRD_OVERVIEW, etc.)
+- Automatic tech stack extraction
+- Code style and testing approach detection
+
+**Character Limits:**
+- Windsurf rules: 12,000 characters per file (enforced via file splitting)
+- Other tools: No enforced limits
+
+**Context Awareness:**
+- Architect mode: References architecture and specification
+- Implementer mode: References implementation details and tasks
+- Reviewer mode: References constitution and quality standards
+- All modes: Full access to framework outputs via file paths
+
+### Breaking Changes
+
+None - fully backward compatible with v0.2.x
+
+### Migration Guide
+
+Existing projects can immediately benefit from the new generators:
+
+1. Run `adf deploy <tool>` in your project directory
+2. Configurations will be generated from existing `.adf/sessions/` data
+3. All tools will recognize and use the new configurations
+4. No manual migration needed
+
+### Next Steps
+
+- **Phase 3**: Enhanced MCP integration for all tools
+- **Phase 4**: Live project monitoring and real-time updates
+- **Phase 5**: Multi-framework project support
+
 ## [0.2.0] - 2025-10-03
 
 ### 🚀 MAJOR RELEASE: Quality-Based Progress Tracking & Resume Capability

package/README.md

@@ -1,6 +1,8 @@
 # @iservu-inc/adf-cli
 
-CLI tool for AgentDevFramework - AI-assisted development framework.
+CLI tool for AgentDevFramework - AI-assisted development framework with multi-provider AI support.
+
+Transform your development workflow with AI-guided requirements gathering using Anthropic Claude, OpenAI GPT, Google Gemini, or OpenRouter.
 
 ## Installation
 
@@ -27,11 +29,12 @@ adf init
 ```
 
 The interactive init process will:
-1. Detect if this is a new or existing project
-2. Ask questions to recommend the best workflow level
-3. Optionally collect documentation URLs (e.g., API docs, design docs)
-4. Optionally collect local documentation file paths (e.g., `./docs/`, `./README.md`)
-5. Offer to deploy to your preferred development tool
+1. Optionally configure AI Provider (or use `adf config` later)
+2. Detect if this is a new or existing project
+3. Ask AI-guided questions to gather requirements
+4. Optionally collect documentation URLs (e.g., API docs, design docs)
+5. Optionally collect local documentation file paths (e.g., `./docs/`, `./README.md`)
+6. Automatically deploy to your preferred development tool
 
 #### Workflow Selection Options
 
@@ -57,6 +60,24 @@ adf init --tool windsurf
 adf init --rapid --tool cursor
 ```
 
+### Configure ADF Settings
+
+Configure ADF settings like AI provider, project preferences, and more:
+
+```bash
+adf config
+```
+
+This command provides an interactive menu with:
+- **AI Provider Setup** - Configure Anthropic, OpenAI, Google Gemini, or OpenRouter
+- Status indicators showing what's configured (green ✓) or not (yellow ○)
+- Future configuration options (coming soon)
+
+You can run `adf config` anytime to:
+- Configure AI for the first time
+- Switch AI providers or models
+- Update your API keys
+
 ### Deploy to Development Tool
 
 Deploy framework configuration to your development tool:
@@ -109,6 +130,70 @@ adf --version
 adf -v
 ```
 
+## AI Provider Configuration
+
+ADF CLI requires an AI provider to guide you through requirements gathering with intelligent follow-up questions and answer quality analysis.
+
+### Supported AI Providers
+
+- **Anthropic Claude** (Claude 3.5 Sonnet, Claude 3 Opus, etc.)
+- **OpenAI** (GPT-4, GPT-4 Turbo, GPT-3.5 Turbo, etc.)
+- **Google Gemini** (Gemini 1.5 Pro, Gemini 1.5 Flash, etc.)
+- **OpenRouter** (Access to 100+ models from multiple providers)
+
+### API Key Setup
+
+Configure your AI provider using the config command:
+
+```bash
+adf config
+```
+
+Then select "AI Provider Setup" to:
+
+1. **Select your AI provider** from the list above
+2. **Enter your API key** (securely saved to `.adf/.env`)
+3. **Choose a model** with type-to-filter autocomplete
+
+Your API key is stored locally in `.adf/.env` and never leaves your machine.
+
+**Note:** You can also configure AI during `adf init`, or skip it and configure later.
+
+**Example `.adf/.env` file:**
+```env
+ANTHROPIC_API_KEY=sk-ant-api03-...
+```
+
+### Getting API Keys
+
+- **Anthropic:** https://console.anthropic.com/
+- **OpenAI:** https://platform.openai.com/api-keys
+- **Google Gemini:** https://ai.google.dev/
+- **OpenRouter:** https://openrouter.ai/keys
+
+### AI-Powered Features
+
+The AI provider enhances your workflow by:
+
+- **Real-Time Answer Quality Analysis** - Scores your answers 0-100
+- **Intelligent Follow-Up Questions** - Automatically generated based on your responses
+- **Context-Aware Guidance** - Tailored suggestions for your project type
+- **Skip Functionality** - Type "skip" anytime to move forward
+
+### Changing Providers
+
+To switch AI providers or models:
+
+```bash
+adf config
+```
+
+Select "AI Provider Setup" and you'll see:
+- ✓ **Configured** status with current provider name
+- Option to reconfigure with a new provider or model
+
+Your previous API keys remain saved in `.adf/.env` for easy switching between providers.
+
 ## What Gets Installed
 
 When you run `adf init`, the following structure is created in your project:
@@ -116,7 +201,9 @@ When you run `adf init`, the following structure is created in your project:
 ```
 your-project/
 ├── .adf/                    # Framework files (isolated)
+│   ├── .env                 # AI provider API keys (gitignored, secure)
 │   ├── context.json         # Your workflow configuration
+│   ├── sessions/            # Requirements gathering sessions
 │   ├── scripts/             # Helper scripts
 │   └── shared/              # Templates, agents, and resources
 │       ├── agents/          # Agent definition files
@@ -246,11 +333,13 @@ When we release updates to the framework:
 
 See [CHANGELOG.md](./CHANGELOG.md) for detailed version history.
 
-**Latest:** v0.1.6
-- Fixed step numbering in init success message
-- Added local documentation files collection
-- Dynamic version tracking in context.json
-- Improved error handling
+**Latest:** v0.3.6
+- AI-guided requirements gathering with multi-provider support
+- Real-time answer quality analysis and intelligent follow-ups
+- Secure API key persistence in `.adf/.env`
+- Dynamic model fetching with autocomplete selection
+- Enhanced UX with visible skip instructions
+- Automatic deployment to your preferred IDE
 
 ## Troubleshooting
 

package/bin/adf.js

@@ -7,6 +7,7 @@ const packageJson = require('../package.json');
 const initCommand = require('../lib/commands/init');
 const deployCommand = require('../lib/commands/deploy');
 const updateCommand = require('../lib/commands/update');
+const configCommand = require('../lib/commands/config');
 
 const program = new Command();
 
@@ -39,6 +40,12 @@ program
   .option('--check', 'Only check for updates, don\'t install')
   .action(updateCommand);
 
+// adf config
+program
+  .command('config')
+  .description('Configure ADF settings (AI provider, etc.)')
+  .action(configCommand);
+
 // Handle unknown commands
 program.on('command:*', () => {
   console.error(chalk.red(`\nInvalid command: ${program.args.join(' ')}`));