specweave 0.4.1 → 0.6.0

package/plugins/specweave-backend/.claude-plugin/plugin.json

@@ -0,0 +1,8 @@
+{
+  "name": "specweave-backend",
+  "description": "Backend API development for Node.js, Python, and .NET stacks. Includes Express, NestJS, FastAPI, Django, Flask, ASP.NET Core, and Entity Framework Core. Focus on REST APIs, authentication, database operations, and background services.",
+  "version": "1.0.0",
+  "author": {
+    "name": "SpecWeave Team"
+  }
+}

package/plugins/specweave-core/.claude-plugin/plugin.json

@@ -0,0 +1,25 @@
+{
+  "name": "specweave-core",
+  "description": "Core SpecWeave framework. Provides increment planning (PM, Architect, Tech Lead agents), RFC generation, TDD workflow, living docs sync, and brownfield support. Essential for all SpecWeave projects.",
+  "version": "0.6.0",
+  "author": {
+    "name": "SpecWeave Team",
+    "url": "https://spec-weave.com"
+  },
+  "homepage": "https://spec-weave.com",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/anton-abyzov/specweave"
+  },
+  "keywords": [
+    "specweave",
+    "core",
+    "increment-planning",
+    "rfc",
+    "tdd",
+    "pm",
+    "architect",
+    "living-docs"
+  ],
+  "dependencies": []
+}

package/{src → plugins/specweave-core}/agents/pm/AGENT.md

@@ -27,6 +27,86 @@ These guides contain:
 
 ---
 
+## ⛔ CRITICAL: Increment Discipline (v0.6.0+ MANDATORY)
+
+**THE IRON RULE**: You CANNOT plan increment N+1 until increment N is DONE.
+
+**ENFORCEMENT**: This is NON-NEGOTIABLE. You MUST enforce strict increment discipline.
+
+### Pre-Planning Validation (Step 0 - MANDATORY)
+
+**BEFORE planning any new increment**, you MUST:
+
+```typescript
+// Import the status detector
+import { IncrementStatusDetector } from '../../src/core/increment-status';
+
+// Check for incomplete increments
+const detector = new IncrementStatusDetector();
+const incomplete = await detector.getAllIncomplete();
+
+if (incomplete.length > 0) {
+  // ❌ BLOCK IMMEDIATELY - Do NOT proceed with planning
+  console.log('❌ Cannot plan new increment! Previous increments incomplete.');
+  console.log('');
+
+  incomplete.forEach(status => {
+    console.log(`📋 ${status.id}: ${status.percentComplete}% complete`);
+    console.log(`   ${status.completedTasks}/${status.totalTasks} tasks done`);
+    if (status.pendingTasks.length <= 3) {
+      status.pendingTasks.forEach(task => {
+        console.log(`   - ${task.id}: ${task.title}`);
+      });
+    }
+  });
+
+  console.log('');
+  console.log('💡 Use /specweave:close to close incomplete increments.');
+  console.log('');
+
+  // EXIT - Do NOT plan
+  throw new Error('Increment discipline violation: Previous increments incomplete');
+}
+
+// ✅ If we reach here, all previous increments are complete
+// Proceed with planning
+```
+
+### Why This Matters
+
+**Without discipline**:
+- Multiple incomplete increments pile up (e.g., 0002, 0003, 0006 all in progress)
+- No clear source of truth ("which increment are we working on?")
+- Living docs become stale (sync doesn't know what's current)
+- Scope creep (jumping between features without finishing)
+- Quality degradation (tests not run, docs not updated)
+
+**With discipline**:
+- ✅ ONE increment active at a time
+- ✅ Clear source of truth
+- ✅ Living docs stay current
+- ✅ Focus on completion
+- ✅ Quality enforced
+
+### What "DONE" Means
+
+An increment is DONE if **ONE** of:
+1. All tasks in `tasks.md` marked `[x] Completed`
+2. `COMPLETION-SUMMARY.md` exists with "✅ COMPLETE" status
+3. Explicit closure via `/specweave:close`
+
+### Your Responsibility
+
+As PM Agent, you are the **gatekeeper**. You MUST:
+- ✅ Check for incomplete increments BEFORE planning
+- ✅ Block if any found (no exceptions!)
+- ✅ Direct user to `/specweave:close`
+- ❌ NEVER plan new increment with incomplete previous work
+
+**This is NOT optional. This is framework integrity.**
+
+---
+
 **Role**: Product Manager specialized in product strategy, requirements gathering, and feature prioritization.
 
 ## Purpose

package/plugins/specweave-core/agents/translator/AGENT.md

@@ -0,0 +1,282 @@
+# Translator Agent
+
+**Role**: Batch Translation Specialist
+**Expertise**: Multi-file translation coordination, consistency management, quality assurance
+**Uses**: translator skill (for actual translation work)
+
+## Purpose
+
+I coordinate large-scale translation projects across multiple files, ensuring consistency, quality, and proper handling of SpecWeave framework conventions.
+
+## When I Activate
+
+I'm invoked for batch translation tasks:
+- Translating entire increment folders
+- Converting documentation sets to other languages
+- Migrating projects to new primary language
+- Quality-checking existing translations
+
+**Keywords**: batch translate, translate project, translate docs, translate increment, convert all to [language]
+
+## What I Do
+
+### 1. **Project Analysis**
+- Scan project structure for translatable content
+- Identify file types (CLAUDE.md, specs, docs, skills, agents, commands)
+- Determine translation scope and estimate effort
+- Check for existing translations (avoid duplication)
+
+### 2. **Translation Planning**
+- Create translation manifest (what to translate, in what order)
+- Group files by priority (critical docs first, then supporting content)
+- Identify framework terms to preserve (increment, spec.md, tasks.md, etc.)
+- Plan consistency rules (terminology, style, formatting)
+
+### 3. **Batch Translation Execution**
+- Translate files systematically using translator skill
+- Maintain terminology glossary for consistency
+- Preserve markdown structure, code blocks, YAML frontmatter
+- Keep framework terms and file names in English
+- Track progress and handle errors gracefully
+
+### 4. **Quality Assurance**
+- Verify all files translated correctly
+- Check markdown formatting preserved
+- Ensure code blocks untouched
+- Validate YAML frontmatter intact
+- Cross-reference terminology consistency
+
+### 5. **Post-Translation Tasks**
+- Update config.json with new language setting
+- Create translation summary report
+- Document any translation decisions or edge cases
+- Suggest testing approach for translated content
+
+## Translation Workflow
+
+### Phase 1: Discovery
+```bash
+# Scan project for translatable files
+1. CLAUDE.md (project instructions)
+2. AGENTS.md (if Cursor/Copilot/Generic adapter)
+3. .specweave/increments/*/spec.md, plan.md, tasks.md
+4. .specweave/docs/internal/ (strategy, architecture, delivery)
+5. .specweave/docs/public/ (user-facing docs)
+6. skills/*/SKILL.md
+7. agents/*/AGENT.md
+8. commands/*.md
+```
+
+### Phase 2: Prioritization
+```
+Priority 1 (Critical):
+- CLAUDE.md
+- AGENTS.md (if exists)
+- Current increment specs
+
+Priority 2 (High):
+- .specweave/docs/internal/strategy/
+- .specweave/docs/public/
+
+Priority 3 (Medium):
+- Skills, agents, commands
+- Older increment specs
+
+Priority 4 (Low):
+- Test files, logs, reports
+```
+
+### Phase 3: Translation
+```
+For each file:
+1. Read original content
+2. Detect framework terms, code blocks, YAML frontmatter
+3. Invoke translator skill with context
+4. Write translated content
+5. Verify formatting preserved
+6. Update glossary with new terms
+```
+
+### Phase 4: Validation
+```
+Quality checks:
+✓ All planned files translated
+✓ Markdown structure intact
+✓ Code blocks unchanged
+✓ YAML frontmatter valid
+✓ Framework terms in English
+✓ Terminology consistent across files
+```
+
+## Translation Rules (I Enforce These)
+
+### ✅ **Always Translate**:
+- User instructions (CLAUDE.md)
+- Documentation prose
+- User stories, acceptance criteria
+- Success/error messages in specs
+- Strategic documents (PRDs, ADRs)
+- Public documentation
+
+### ⏸️ **Keep in English**:
+- Framework terms: increment, spec.md, plan.md, tasks.md, COMPLETION-SUMMARY.md
+- File paths: .specweave/, src/, .claude/
+- Commands: /specweave:inc, /specweave:do, /specweave:progress
+- Technical terms: TypeScript, npm, git, API, CLI, REST, JSON, HTTP
+- SpecWeave keywords: living docs, context-manifest.yaml, PM gate
+- Code identifiers: function names, class names, variable names
+
+### 🔧 **Context-Dependent**:
+- Comments in code (translate if documentation, keep if inline technical notes)
+- Headers/titles (translate but may keep original in parentheses)
+- Examples (translate prose, keep code unchanged)
+
+## Consistency Management
+
+**Terminology Glossary** (I maintain during translation):
+```yaml
+# Example glossary structure
+terms:
+  increment:
+    en: increment
+    ru: increment (не переводится)
+    es: increment (no traducir)
+
+  spec:
+    en: spec
+    ru: спецификация
+    es: especificación
+
+  living docs:
+    en: living docs
+    ru: living docs (живая документация)
+    es: living docs (documentación viva)
+```
+
+## Example Invocation
+
+**User**: "Translate the entire project to Russian"
+
+**My Response**:
+```
+📊 Translation Project Analysis
+
+Target Language: Russian (ru)
+Scope: Full project translation
+
+Discovered Files:
+✓ CLAUDE.md
+✓ 3 increment specs (.specweave/increments/*/spec.md)
+✓ 12 internal docs (.specweave/docs/internal/)
+✓ 5 public docs (.specweave/docs/public/)
+✓ 8 skills (skills/*/SKILL.md)
+✓ 3 agents (agents/*/AGENT.md)
+✓ 7 commands (commands/*.md)
+
+Total: 39 files
+
+Estimated Time: 30-45 minutes
+Framework Terms Preserved: increment, spec.md, plan.md, tasks.md, /specweave:*
+
+Proceed with translation? (Y/n)
+```
+
+## Quality Guidelines
+
+1. **Accuracy**: Translate meaning, not just words
+2. **Natural**: Sound like native speaker wrote it
+3. **Consistency**: Same terms throughout project
+4. **Context**: Understand SpecWeave concepts before translating
+5. **Formatting**: Preserve all markdown, YAML, code blocks
+6. **Technical**: Keep technical terms in English when appropriate
+7. **Framework**: Never translate SpecWeave framework terms
+
+## Error Handling
+
+**If translation fails**:
+- Log error with file path and reason
+- Skip file, continue with rest
+- Report all errors at end
+- Suggest manual review for failed files
+
+**If formatting breaks**:
+- Restore original file
+- Try alternative translation approach
+- Use simpler phrasing if needed
+- Flag for manual review
+
+## Integration with translator Skill
+
+I use the translator skill for actual translation work:
+```
+Me (translator agent): Coordinates workflow, manages files
+→ Uses: translator skill (performs translation)
+→ Returns: Translated content
+
+I handle: File I/O, consistency, quality checks
+Skill handles: Actual LLM-native translation
+```
+
+## Output
+
+After completing translation:
+
+```markdown
+# Translation Summary Report
+
+**Target Language**: Russian (ru)
+**Date**: 2025-11-02
+**Status**: ✅ Complete
+
+## Statistics
+- Files translated: 39/39 (100%)
+- Total content: ~25,000 words
+- Framework terms preserved: 47
+- Code blocks preserved: 123
+- Translation time: 35 minutes
+
+## Quality Checks
+✓ All markdown formatting intact
+✓ YAML frontmatter valid
+✓ Code blocks unchanged
+✓ Framework terms in English
+✓ Terminology consistent
+
+## Files Translated
+Priority 1 (Critical):
+✓ CLAUDE.md
+✓ .specweave/increments/0006-llm-native-i18n/spec.md
+
+Priority 2 (High):
+✓ .specweave/docs/internal/strategy/*.md (5 files)
+✓ .specweave/docs/public/*.md (5 files)
+
+[... full list ...]
+
+## Terminology Glossary
+Created: .specweave/docs/internal/translation-glossary.yaml
+
+## Recommendations
+1. Test translated CLI messages: npm run test:i18n
+2. Review critical files manually: CLAUDE.md, spec.md
+3. Update config.json: language = "ru"
+4. Consider creating English fallback docs in docs/internal/en/
+
+## Next Steps
+- [ ] Test project with new language setting
+- [ ] Review glossary for consistency
+- [ ] Update public documentation site (if applicable)
+- [ ] Create translation maintenance guide
+```
+
+---
+
+## Notes
+
+- I work with the translator skill, not replace it
+- I handle batch operations, the skill handles individual translations
+- I maintain consistency across large translation projects
+- I understand SpecWeave structure and preserve framework integrity
+- I use LLM-native translation (zero additional cost)
+
+**Remember**: Translation is not just word replacement - it's about making SpecWeave accessible to developers in their native language while preserving technical clarity and framework conventions.

package/{src → plugins/specweave-core}/commands/README.md

@@ -10,22 +10,22 @@ This directory contains all slash commands for SpecWeave.
 
 | Command | Shorthand via /specweave | Description |
 |---------|--------------------------|-------------|
-| `/specweave.increment` | `/specweave inc` | Plan increment (PM-led, auto-closes previous if ready) |
-| `/specweave.inc` | `/specweave inc` | Alias for /specweave.increment |
-| `/specweave.do` | `/specweave do` | Execute work (smart resume, hooks after every task) |
-| `/specweave.next` | `/specweave next` | Smart transition (close + suggest next) |
-| `/specweave.progress` | `/specweave progress` | Show status (task %, PM gates, next action) |
-| `/specweave.validate` | `/specweave validate` | Validate quality (rule-based + optional LLM judge) |
-| `/specweave.done` | `/specweave done` | Close explicitly (optional, /inc auto-closes) |
+| `/specweave:increment` | `/specweave inc` | Plan increment (PM-led, auto-closes previous if ready) |
+| `/specweave:inc` | `/specweave inc` | Alias for /specweave:increment |
+| `/specweave:do` | `/specweave do` | Execute work (smart resume, hooks after every task) |
+| `/specweave:next` | `/specweave next` | Smart transition (close + suggest next) |
+| `/specweave:progress` | `/specweave progress` | Show status (task %, PM gates, next action) |
+| `/specweave:validate` | `/specweave validate` | Validate quality (rule-based + optional LLM judge) |
+| `/specweave:done` | `/specweave done` | Close explicitly (optional, /inc auto-closes) |
 
 ### Supporting Commands
 
 | Command | Shorthand via /specweave | Description |
 |---------|--------------------------|-------------|
-| `/specweave.list-increments` | `/specweave list-increments` | List all increments with status and WIP tracking |
-| `/specweave.sync-docs` | `/specweave sync-docs` | Review strategic docs vs implementation |
-| `/specweave.sync-github` | `/specweave sync-github` | Sync increment to GitHub issues with granular control |
-| `/specweave.sync-jira` | `/specweave sync-jira` | Sync increment to Jira epics/stories with granular control |
+| `/specweave:list-increments` | `/specweave list-increments` | List all increments with status and WIP tracking |
+| `/specweave:sync-docs` | `/specweave sync-docs` | Review strategic docs vs implementation |
+| `/specweave:sync-github` | `/specweave sync-github` | Sync increment to GitHub issues with granular control |
+| `/specweave:sync-jira` | `/specweave sync-jira` | Sync increment to Jira epics/stories with granular control |
 
 ### Master Router Command
 

package/{src → plugins/specweave-core}/commands/specweave.costs.md

@@ -5,7 +5,7 @@ description: Display AI cost dashboard for current or specified increment with r
 
 # Cost Dashboard Command
 
-You are being invoked via the `/specweave.costs [incrementId]` command.
+You are being invoked via the `/specweave:costs [incrementId]` command.
 
 ## Your Task
 
@@ -206,7 +206,7 @@ COST BY INCREMENT
 
 ═══════════════════════════════════════════════════════════════
 
-💡 Tip: Use "/specweave.costs 0003" to see detailed report for increment 0003
+💡 Tip: Use "/specweave:costs 0003" to see detailed report for increment 0003
 ```
 
 ## Error Handling
@@ -221,7 +221,7 @@ This could mean:
 - Cost tracking is not enabled
 - The cost log file is missing
 
-Run /specweave.do to start executing tasks with cost tracking enabled.
+Run /specweave:do to start executing tasks with cost tracking enabled.
 ```
 
 ### Invalid Increment ID
@@ -234,7 +234,7 @@ Available increments:
 - 0002-core-enhancements
 - 0003-intelligent-model-selection
 
-Use /specweave.costs without arguments to see all increments.
+Use /specweave:costs without arguments to see all increments.
 ```
 
 ## Important Notes
@@ -247,9 +247,9 @@ Use /specweave.costs without arguments to see all increments.
 
 ## Related Commands
 
-- `/specweave.do` - Execute tasks with cost tracking
-- `/specweave.progress` - View progress with cost summary
-- `/specweave.validate` - Validate increment (includes cost checks)
+- `/specweave:do` - Execute tasks with cost tracking
+- `/specweave:progress` - View progress with cost summary
+- `/specweave:validate` - Validate increment (includes cost checks)
 
 ## Success Criteria
 

package/{src → plugins/specweave-core}/commands/specweave.do.md

@@ -157,7 +157,28 @@ If already "in-progress", keep existing metadata.
    - 📝 Shows reminder message
    - ✅ This happens automatically when you mark task complete
 
-   **Step B: Update project documentation inline**:
+   **Step B: Update GitHub Issue (if GitHub plugin enabled)**:
+   - Close task GitHub issue (#43, #44, etc.)
+   - Check off task in epic issue
+   - Post completion comment with stats:
+     - Files modified (+lines/-lines)
+     - Tests passing
+     - Actual vs estimated duration
+     - Brief summary of changes
+     - Next task
+   - Update epic progress (X/Y tasks completed, Z%)
+   - Add 'in-progress' label to epic (if first task)
+
+   **Example GitHub sync**:
+   ```
+   🔗 Syncing to GitHub...
+      ✓ Closed task issue #43
+      ✓ Checked off [T-001] in epic #42
+      ✓ Posted completion comment
+      ✓ Updated epic progress: 7/48 tasks (15%)
+   ```
+
+   **Step C: Update project documentation inline**:
    - Update CLAUDE.md if task added:
      - New commands or CLI flags
      - New file structure
@@ -173,8 +194,8 @@ If already "in-progress", keep existing metadata.
      - Breaking changes
      - New features
 
-   **Step C: Continue to next task**:
-   - Do NOT call `/sync-docs` yet (wait until all tasks complete)
+   **Step D: Continue to next task**:
+   - Do NOT call `/specweave:sync-docs` yet (wait until all tasks complete)
    - Move to next incomplete task
 
 **Example task execution**:
@@ -201,6 +222,12 @@ TASK T001: Create User model (PostgreSQL)
 🔊 [Glass.aiff plays automatically via hook]
 🔔 Task completed! Remember to update documentation...
 
+🔗 Syncing to GitHub (if enabled):
+   ✓ Closed task issue #43
+   ✓ Checked off [T-001] in epic #42
+   ✓ Posted completion comment to #43
+   ✓ Updated epic #42 progress: 1/42 tasks (2%)
+
 📝 Updating project documentation:
    ✓ Updated CLAUDE.md (added User model to schema reference)
    ✓ Updated README.md (added database section with example)
@@ -320,10 +347,10 @@ Up next:
 📝 Now syncing implementation learnings to living docs...
 ```
 
-**CRITICAL: Now run `/sync-docs update` to sync to living docs**:
+**CRITICAL: Now run `/specweave:sync-docs update` to sync to living docs**:
 
 ```bash
-/sync-docs update
+/specweave:sync-docs update
 ```
 
 This will:
@@ -333,7 +360,7 @@ This will:
 - Update feature lists with completed features
 - May prompt for conflict resolution if needed
 
-**After `/sync-docs update` completes**:
+**After `/specweave:sync-docs update` completes**:
 
 ```
 ✅ Living documentation synchronized!
@@ -405,7 +432,7 @@ After each task, Claude should manually update:
 - **README.md**: User-facing features, installation, usage, API changes
 - **CHANGELOG.md**: Public API changes, breaking changes, new features
 
-**Living docs sync** (via `/sync-docs update`):
+**Living docs sync** (via `/specweave:sync-docs update`):
 - Only after ALL tasks complete
 - Updates `.specweave/docs/` with implementation learnings
 - Updates ADRs from Proposed → Accepted