proagents 1.0.0

package/proagents/getting-started/ai-training-setup.md

@@ -0,0 +1,380 @@
+# Getting Started with AI Training
+
+Configure the learning system to improve ProAgents for your project.
+
+---
+
+## What Does AI Training Do?
+
+The learning system helps ProAgents:
+
+1. **Learn your patterns** - Naming conventions, code style, architecture
+2. **Remember corrections** - Auto-apply fixes you've made before
+3. **Adapt to preferences** - Skip checkpoints you always skip
+4. **Improve suggestions** - Better code generation over time
+
+---
+
+## Quick Start
+
+### Step 1: Enable Learning
+
+```yaml
+# proagents.config.yaml
+learning:
+  enabled: true
+```
+
+That's it! Learning starts automatically.
+
+### Step 2: Use ProAgents Normally
+
+As you work, the system learns:
+
+```bash
+/feature-start "Add user profile"
+# → Learns feature naming patterns
+
+/analyze
+# → Learns project structure
+
+/implement
+# → Learns code patterns
+```
+
+### Step 3: Check Learning Progress
+
+```bash
+/learning status
+```
+
+Output:
+```
+Learning System Status
+══════════════════════════════════════
+
+Patterns Learned: 45
+├── Naming conventions: 12 (high confidence)
+├── Code style: 18 (high confidence)
+├── Architecture: 8 (medium confidence)
+└── Testing: 7 (learning)
+
+Corrections Applied: 23
+User Preferences: 8
+
+Last Updated: 2 hours ago
+```
+
+---
+
+## What Gets Learned
+
+### 1. Code Patterns
+
+```
+✓ Component structure patterns
+✓ Hook implementation patterns
+✓ Service layer patterns
+✓ Import ordering
+✓ Error handling approaches
+✓ Testing patterns
+```
+
+### 2. Naming Conventions
+
+```
+✓ File naming (PascalCase, kebab-case, etc.)
+✓ Function naming
+✓ Variable naming
+✓ Type/Interface naming
+✓ Test naming
+```
+
+### 3. User Preferences
+
+```
+✓ Checkpoint skipping patterns
+✓ Documentation level preferences
+✓ Review thoroughness
+✓ Common corrections
+```
+
+### 4. Project Domain
+
+```
+✓ Business terminology
+✓ Domain entities
+✓ Specific patterns for your domain
+```
+
+---
+
+## Configuration Options
+
+### Basic Configuration
+
+```yaml
+learning:
+  enabled: true
+  track_preferences: true
+  track_patterns: true
+  track_corrections: true
+```
+
+### Advanced Configuration
+
+```yaml
+learning:
+  enabled: true
+
+  # What to learn
+  tracking:
+    preferences: true        # User checkpoint/workflow preferences
+    patterns: true           # Code patterns and conventions
+    corrections: true        # Manual corrections to AI output
+    domain: true             # Domain-specific terminology
+
+  # Auto-apply settings
+  auto_apply:
+    corrections: true        # Auto-apply learned corrections
+    threshold: 3             # Apply after N occurrences
+
+  # Privacy settings
+  privacy:
+    share_globally: false    # Keep learnings project-local
+    anonymize: true          # Anonymize metrics
+    exclude_secrets: true    # Never learn from secret values
+
+  # Storage
+  storage:
+    location: ".proagents/.learning"
+    max_size: "50MB"
+    retention_days: 365
+
+  # Performance
+  performance:
+    batch_updates: true
+    update_frequency: "on_commit"  # on_commit, hourly, daily
+```
+
+---
+
+## Teaching the System
+
+### Explicit Teaching
+
+You can explicitly teach patterns:
+
+```bash
+# Teach a naming convention
+/learning teach "Components should use PascalCase with 'View' suffix"
+
+# Teach a code pattern
+/learning teach "Always use React Query for data fetching"
+
+# Teach domain terminology
+/learning teach "A 'Workspace' contains multiple 'Projects'"
+```
+
+### Correction Learning
+
+When you correct AI output, the system learns:
+
+**AI suggests:**
+```typescript
+const fetchData = async () => {
+  const response = await fetch('/api/users');
+  return response.json();
+}
+```
+
+**You change to:**
+```typescript
+const fetchData = async () => {
+  const response = await apiClient.get('/users');
+  return response.data;
+}
+```
+
+**System learns:**
+```
+Pattern: Use apiClient instead of fetch
+Confidence: Low (1 occurrence)
+```
+
+After 3+ similar corrections:
+```
+Pattern: Use apiClient instead of fetch
+Confidence: High
+Auto-apply: Yes
+```
+
+---
+
+## Training Data Sources
+
+### Automatic Sources
+
+| Source | What's Learned |
+|--------|----------------|
+| Codebase | Patterns, style, architecture |
+| Git history | Change patterns, frequent edits |
+| PR comments | Review feedback patterns |
+| Corrections | User preferences |
+
+### Manual Sources
+
+```yaml
+learning:
+  sources:
+    # Explicit documentation
+    - path: "./docs/coding-standards.md"
+      type: "documentation"
+      weight: 1.0
+
+    # Domain knowledge
+    - path: "./docs/domain-model.md"
+      type: "domain"
+      weight: 0.8
+
+    # Example code
+    - path: "./examples/"
+      type: "examples"
+      weight: 0.9
+```
+
+---
+
+## Learning Reports
+
+### View Learning Summary
+
+```bash
+/learning report
+```
+
+Output:
+```
+┌─────────────────────────────────────────────────────────┐
+│ Learning Report - January 2024                          │
+├─────────────────────────────────────────────────────────┤
+│ Patterns Learned: 45                                    │
+│ Auto-corrections Applied: 23                            │
+│ Checkpoints Auto-skipped: 36 (saved ~3 hrs)            │
+│                                                         │
+│ Top Improvements:                                       │
+│ • Now auto-detects auth context (saved 15min/feature)  │
+│ • Learned preferred test structure (fewer corrections) │
+│ • Optimized checkpoint flow for this project           │
+├─────────────────────────────────────────────────────────┤
+│ Suggestions:                                            │
+│ • Consider documenting the custom hook pattern         │
+│ • Test coverage trending down (now 72%, was 80%)       │
+└─────────────────────────────────────────────────────────┘
+```
+
+### View Specific Patterns
+
+```bash
+/learning patterns           # View all learned patterns
+/learning patterns naming    # View naming patterns
+/learning patterns testing   # View testing patterns
+```
+
+### View Corrections
+
+```bash
+/learning corrections        # View all auto-corrections
+/learning corrections active # View active corrections
+```
+
+---
+
+## Managing Learning Data
+
+### Export Learning Data
+
+```bash
+/learning export --output ./learning-export.json
+```
+
+### Import Learning Data
+
+```bash
+/learning import ./learning-export.json
+```
+
+### Reset Learning
+
+```bash
+# Reset specific category
+/learning reset patterns
+
+# Reset all learning
+/learning reset --all
+
+# Reset with confirmation
+/learning reset --all --confirm
+```
+
+### Disable Specific Learning
+
+```bash
+# Temporarily disable a learned pattern
+/learning disable "Use apiClient instead of fetch"
+
+# Re-enable
+/learning enable "Use apiClient instead of fetch"
+```
+
+---
+
+## Troubleshooting
+
+### Learning Not Working
+
+```bash
+# Check status
+/learning status
+
+# Debug mode
+/learning debug
+```
+
+### Wrong Pattern Learned
+
+```bash
+# Remove incorrect pattern
+/learning remove "incorrect pattern"
+
+# Or correct it
+/learning correct "old pattern" --to "correct pattern"
+```
+
+### Performance Issues
+
+```yaml
+learning:
+  performance:
+    batch_updates: true
+    update_frequency: "daily"  # Reduce frequency
+    max_patterns: 100          # Limit stored patterns
+```
+
+---
+
+## Best Practices
+
+1. **Be consistent** - Consistent code leads to confident patterns
+2. **Correct promptly** - Correct AI output to teach preferences
+3. **Document standards** - Explicit docs help learning
+4. **Review periodically** - Check learned patterns monthly
+5. **Export before reset** - Backup learning data
+
+---
+
+## Next Steps
+
+- [Configure IDE](./ide-setup.md)
+- [Set up standards](../standards/)
+- [Review AI training details](../ai-training/)

package/proagents/getting-started/ide-setup.md

@@ -0,0 +1,195 @@
+# IDE Setup Guide
+
+Quick setup guides for configuring your IDE with ProAgents.
+
+---
+
+## Supported IDEs
+
+| IDE | Guide | Status |
+|-----|-------|--------|
+| VS Code | [Full Guide](../team/ide-setup/vscode.md) | Recommended |
+| JetBrains | [Full Guide](../team/ide-setup/jetbrains.md) | Supported |
+| Cursor | [Full Guide](../team/ide-setup/cursor.md) | Supported |
+| Neovim | [Full Guide](../team/ide-setup/neovim.md) | Supported |
+
+---
+
+## Quick Setup
+
+### VS Code
+
+1. **Install Extensions**
+```bash
+# Essential
+code --install-extension dbaeumer.vscode-eslint
+code --install-extension esbenp.prettier-vscode
+
+# AI Integration (choose one)
+code --install-extension GitHub.copilot
+code --install-extension Continue.continue
+```
+
+2. **Configure Settings**
+
+Create `.vscode/settings.json`:
+```json
+{
+  "editor.formatOnSave": true,
+  "editor.defaultFormatter": "esbenp.prettier-vscode",
+  "editor.codeActionsOnSave": {
+    "source.fixAll.eslint": "explicit"
+  }
+}
+```
+
+3. **Set Up Tasks**
+
+Create `.vscode/tasks.json`:
+```json
+{
+  "version": "2.0.0",
+  "tasks": [
+    {
+      "label": "ProAgents: Start Feature",
+      "type": "shell",
+      "command": "proagents feature start"
+    },
+    {
+      "label": "ProAgents: Run Tests",
+      "type": "shell",
+      "command": "proagents test"
+    }
+  ]
+}
+```
+
+---
+
+### JetBrains (WebStorm/IntelliJ)
+
+1. **Install Plugins**
+   - Settings → Plugins → Marketplace
+   - Search and install: ESLint, Prettier
+
+2. **Configure ESLint**
+   - Settings → Languages & Frameworks → JavaScript → Code Quality Tools → ESLint
+   - Enable "Automatic ESLint configuration"
+
+3. **Configure Prettier**
+   - Settings → Languages & Frameworks → JavaScript → Prettier
+   - Enable "Run on save"
+
+---
+
+### Cursor
+
+Cursor comes with AI features built-in. Additional setup:
+
+1. **Configure AI Context**
+```
+Settings → AI → Context
+Add: proagents/**/*.md
+```
+
+2. **Install Extensions**
+   - Same as VS Code (Cursor supports VS Code extensions)
+
+---
+
+### Neovim
+
+1. **Install LSP Support**
+```lua
+-- Using lazy.nvim
+{
+  'neovim/nvim-lspconfig',
+  dependencies = {
+    'williamboman/mason.nvim',
+    'williamboman/mason-lspconfig.nvim',
+  },
+}
+```
+
+2. **Configure TypeScript**
+```lua
+require('lspconfig').tsserver.setup({})
+require('lspconfig').eslint.setup({})
+```
+
+---
+
+## AI Tool Integration
+
+### GitHub Copilot
+- Works with VS Code, JetBrains, Neovim
+- Provides inline code suggestions
+- No special ProAgents configuration needed
+
+### Claude Code
+- CLI-based, works with any editor
+- Run `claude-code` in terminal alongside your IDE
+
+### Continue
+- VS Code extension
+- Configure custom context with ProAgents prompts
+
+### Codeium
+- Free alternative to Copilot
+- Works with most editors
+
+---
+
+## Common Tasks
+
+### Setting Up Keyboard Shortcuts
+
+**VS Code:**
+```json
+// keybindings.json
+[
+  {
+    "key": "ctrl+shift+p f",
+    "command": "workbench.action.terminal.sendSequence",
+    "args": { "text": "proagents feature start\n" }
+  }
+]
+```
+
+### Terminal Integration
+
+Most ProAgents commands run in the terminal. Configure your IDE's integrated terminal:
+
+**VS Code:**
+```json
+{
+  "terminal.integrated.defaultProfile.linux": "bash",
+  "terminal.integrated.defaultProfile.osx": "zsh"
+}
+```
+
+---
+
+## Troubleshooting
+
+### ESLint Not Working
+1. Ensure `node_modules` is installed
+2. Check ESLint extension is enabled
+3. Verify `.eslintrc` exists
+
+### Prettier Conflicts
+1. Disable other formatters
+2. Set Prettier as default formatter
+3. Check for conflicting settings
+
+### TypeScript Errors
+1. Restart TS server: `Cmd/Ctrl+Shift+P` → "TypeScript: Restart TS Server"
+2. Ensure correct TypeScript version
+
+---
+
+## Related Resources
+
+- [VS Code Full Guide](../team/ide-setup/vscode.md)
+- [JetBrains Full Guide](../team/ide-setup/jetbrains.md)
+- [Team Onboarding](./team-onboarding.md)