clavix 4.1.0 → 4.1.2

package/dist/cli/commands/update.js

@@ -242,31 +242,8 @@ clavix start
 Learn more: https://github.com/clavixdev/clavix`;
     }
     getClaudeContent() {
-        return `## Clavix Integration
-
-This project uses Clavix for prompt improvement and PRD generation. The following slash commands are available:
-
-### /clavix:fast [prompt]
-Quick prompt improvements with smart triage. Clavix will analyze your prompt and recommend deep analysis if needed. Perfect for making "shitty prompts good" quickly.
-
-### /clavix:deep [prompt]
-Comprehensive prompt analysis with alternative phrasings, edge cases, implementation examples, and potential issues. Use for complex requirements or when you want thorough exploration.
-
-### /clavix:prd
-Launch the PRD generation workflow. Clavix will guide you through strategic questions and generate both a comprehensive PRD and a quick-reference version optimized for AI consumption.
-
-### /clavix:start
-Enter conversational mode for iterative prompt development. Discuss your requirements naturally, and later use \`/clavix:summarize\` to extract an optimized prompt.
-
-### /clavix:summarize
-Analyze the current conversation and extract key requirements into a structured prompt and mini-PRD.
-
-**When to use which mode:**
-- **Fast mode** (\`/clavix:fast\`): Quick cleanup for simple prompts
-- **Deep mode** (\`/clavix:deep\`): Comprehensive analysis for complex requirements
-- **PRD mode** (\`/clavix:prd\`): Strategic planning with architecture and business impact
-
-**Pro tip**: Start complex features with \`/clavix:prd\` or \`/clavix:start\` to ensure clear requirements before implementation.`;
+        // Use DocInjector as single source of truth for CLAUDE.md content
+        return DocInjector.getClaudeBlockContent();
     }
     hasUpToDateBlock(currentContent, newContent) {
         // Check if the managed block contains the new content

package/dist/core/doc-injector.d.ts

@@ -43,6 +43,11 @@ export declare class DocInjector {
      * Create default AGENTS.md content
      */
     static getDefaultAgentsContent(): string;
+    /**
+     * Create the CLAUDE.md block content (without file wrapper)
+     * This is the single source of truth for Claude Code documentation
+     */
+    static getClaudeBlockContent(): string;
     /**
      * Create default CLAUDE.md content for Claude Code
      */

package/dist/core/doc-injector.js

@@ -161,37 +161,73 @@ For more information, run \`clavix --help\` in your terminal.
 `;
     }
     /**
-     * Create default CLAUDE.md content for Claude Code
+     * Create the CLAUDE.md block content (without file wrapper)
+     * This is the single source of truth for Claude Code documentation
      */
-    static getDefaultClaudeContent() {
-        return `# Claude Code Instructions
-
-<!-- CLAVIX:START -->
-## Clavix Integration
+    static getClaudeBlockContent() {
+        return `## Clavix Integration
 
 This project uses Clavix for prompt improvement and PRD generation. The following slash commands are available:
 
-### /clavix:fast [prompt]
+### Prompt Optimization Commands
+
+#### /clavix:fast [prompt]
 Quick prompt improvements with smart triage. Clavix will analyze your prompt and recommend deep analysis if needed. Perfect for making "shitty prompts good" quickly.
 
-### /clavix:deep [prompt]
+#### /clavix:deep [prompt]
 Comprehensive prompt analysis with alternative phrasings, edge cases, implementation examples, and potential issues. Use for complex requirements or when you want thorough exploration.
 
-### /clavix:prd
+### PRD & Planning Commands
+
+#### /clavix:prd
 Launch the PRD generation workflow. Clavix will guide you through strategic questions and generate both a comprehensive PRD and a quick-reference version optimized for AI consumption.
 
-### /clavix:start
+#### /clavix:plan
+Generate an optimized implementation task breakdown from your PRD. Creates a phased task plan with dependencies and priorities.
+
+#### /clavix:implement
+Execute tasks from your task plan with AI assistance. Supports automatic git commits and progress tracking.
+
+### Session Management Commands
+
+#### /clavix:start
 Enter conversational mode for iterative prompt development. Discuss your requirements naturally, and later use \`/clavix:summarize\` to extract an optimized prompt.
 
-### /clavix:summarize
+#### /clavix:summarize
 Analyze the current conversation and extract key requirements into a structured prompt and mini-PRD.
 
+### Utility Commands
+
+#### /clavix:execute
+Run saved prompts with lifecycle awareness. Execute previously saved fast/deep prompts.
+
+#### /clavix:prompts
+Manage your saved fast/deep prompts. List, view, and organize your prompt library.
+
+#### /clavix:archive
+Archive completed projects. Move finished PRDs and outputs to the archive for future reference.
+
 **When to use which mode:**
 - **Fast mode** (\`/clavix:fast\`): Quick cleanup for simple prompts
 - **Deep mode** (\`/clavix:deep\`): Comprehensive analysis for complex requirements
 - **PRD mode** (\`/clavix:prd\`): Strategic planning with architecture and business impact
 
-**Pro tip**: Start complex features with \`/clavix:prd\` or \`/clavix:start\` to ensure clear requirements before implementation.
+**Recommended Workflow:**
+1. Start with \`/clavix:prd\` or \`/clavix:start\` for complex features
+2. Generate tasks with \`/clavix:plan\`
+3. Implement with \`/clavix:implement\`
+4. Archive when complete with \`/clavix:archive\`
+
+**Pro tip**: Start complex features with \`/clavix:prd\` or \`/clavix:start\` to ensure clear requirements before implementation.`;
+    }
+    /**
+     * Create default CLAUDE.md content for Claude Code
+     */
+    static getDefaultClaudeContent() {
+        return `# Claude Code Instructions
+
+<!-- CLAVIX:START -->
+${this.getClaudeBlockContent()}
 <!-- CLAVIX:END -->
 `;
     }

package/dist/templates/slash-commands/_canonical/deep.md

@@ -9,11 +9,41 @@ You are helping the user perform comprehensive deep analysis using Clavix Intell
 
 ---
 
-{{INCLUDE:mode-headers/planning-mode.md MODE="deep analysis"}}
+## CLAVIX MODE: Requirements & Planning Only
+
+**You are in Clavix deep analysis mode. You help perform comprehensive prompt analysis, NOT implement features.**
+
+**YOUR ROLE:**
+- ✓ Analyze prompts for quality
+- ✓ Apply all optimization patterns
+- ✓ Generate alternative approaches
+- ✓ Identify edge cases and validation checklists
+- ✓ Provide comprehensive quality assessments
+
+**DO NOT IMPLEMENT. DO NOT IMPLEMENT. DO NOT IMPLEMENT.**
+- ✗ DO NOT write application code for the feature
+- ✗ DO NOT implement what the prompt/PRD describes
+- ✗ DO NOT generate actual components/functions
+
+**You are analyzing prompts, not building what they describe.**
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 
-{{INCLUDE:mode-headers/self-correction.md WORKFLOW="deep prompt analysis"}}
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself:
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+
+**STOP**: Immediately halt code generation
+
+**CORRECT**: Output:
+"I apologize - I was implementing instead of analyzing. Let me return to deep prompt analysis."
+
+**RESUME**: Return to the deep prompt analysis workflow.
 
 ---
 

package/dist/templates/slash-commands/_canonical/execute.md

@@ -11,7 +11,29 @@ Prompts are automatically saved to `.clavix/outputs/prompts/fast/` or `.clavix/o
 
 ---
 
-{{INCLUDE:mode-headers/implementation-mode.md SOURCE="prompts"}}
+## CLAVIX MODE: Implementation
+
+**You are in Clavix implementation mode. You ARE authorized to write code and implement features.**
+
+**YOUR ROLE:**
+- ✓ Read and understand prompt requirements
+- ✓ Implement the optimized prompt
+- ✓ Write production-quality code
+- ✓ Follow prompt specifications
+- ✓ Execute saved prompts from fast/deep modes
+
+**IMPLEMENTATION AUTHORIZED:**
+- ✓ Writing functions, classes, and components
+- ✓ Creating new files and modifying existing ones
+- ✓ Implementing features described in saved prompts
+- ✓ Writing tests for implemented code
+
+**MODE ENTRY VALIDATION:**
+Before implementing, verify:
+1. Source documents exist (prompts in .clavix/outputs/prompts/)
+2. Output assertion: "Entering IMPLEMENTATION mode. I will implement the saved prompt."
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 

package/dist/templates/slash-commands/_canonical/fast.md

@@ -9,11 +9,40 @@ You are helping the user improve their prompt using Clavix's fast mode, which ap
 
 ---
 
-{{INCLUDE:mode-headers/planning-mode.md MODE="prompt optimization"}}
+## CLAVIX MODE: Requirements & Planning Only
+
+**You are in Clavix prompt optimization mode. You help analyze and optimize PROMPTS, NOT implement features.**
+
+**YOUR ROLE:**
+- ✓ Analyze prompts for quality
+- ✓ Apply optimization patterns
+- ✓ Generate improved versions
+- ✓ Provide quality assessments
+
+**DO NOT IMPLEMENT. DO NOT IMPLEMENT. DO NOT IMPLEMENT.**
+- ✗ DO NOT write application code for the feature
+- ✗ DO NOT implement what the prompt/PRD describes
+- ✗ DO NOT generate actual components/functions
+
+**You are optimizing prompts, not building what they describe.**
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 
-{{INCLUDE:mode-headers/self-correction.md WORKFLOW="prompt optimization"}}
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself:
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+
+**STOP**: Immediately halt code generation
+
+**CORRECT**: Output:
+"I apologize - I was implementing instead of optimizing the prompt. Let me return to prompt optimization."
+
+**RESUME**: Return to the prompt optimization workflow.
 
 ---
 

package/dist/templates/slash-commands/_canonical/implement.md

@@ -9,7 +9,29 @@ You are helping the user implement tasks from their task plan with AI assistance
 
 ---
 
-{{INCLUDE:mode-headers/implementation-mode.md SOURCE="tasks"}}
+## CLAVIX MODE: Implementation
+
+**You are in Clavix implementation mode. You ARE authorized to write code and implement features.**
+
+**YOUR ROLE:**
+- ✓ Read and understand task requirements
+- ✓ Implement tasks from tasks.md
+- ✓ Write production-quality code
+- ✓ Follow PRD specifications
+- ✓ Run `clavix task-complete` after each task
+
+**IMPLEMENTATION AUTHORIZED:**
+- ✓ Writing functions, classes, and components
+- ✓ Creating new files and modifying existing ones
+- ✓ Implementing features described in tasks.md
+- ✓ Writing tests for implemented code
+
+**MODE ENTRY VALIDATION:**
+Before implementing, verify:
+1. Source documents exist (tasks.md in .clavix/outputs/)
+2. Output assertion: "Entering IMPLEMENTATION mode. I will implement tasks from tasks.md."
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 

package/dist/templates/slash-commands/_canonical/plan.md

@@ -9,11 +9,41 @@ You are helping the user generate an optimized implementation task breakdown fro
 
 ---
 
-{{INCLUDE:mode-headers/planning-mode.md MODE="task breakdown"}}
+## CLAVIX MODE: Requirements & Planning Only
+
+**You are in Clavix task breakdown mode. You help generate implementation tasks from PRDs, NOT implement features.**
+
+**YOUR ROLE:**
+- ✓ Read and analyze PRD documents
+- ✓ Generate structured task breakdowns
+- ✓ Organize tasks into logical phases
+- ✓ Create atomic, actionable task descriptions
+- ✓ Assign task IDs and references
+
+**DO NOT IMPLEMENT. DO NOT IMPLEMENT. DO NOT IMPLEMENT.**
+- ✗ DO NOT write application code for the feature
+- ✗ DO NOT implement what the PRD describes
+- ✗ DO NOT generate actual components/functions
+
+**You are generating tasks, not implementing them.**
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 
-{{INCLUDE:mode-headers/self-correction.md WORKFLOW="task breakdown generation"}}
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself:
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+
+**STOP**: Immediately halt code generation
+
+**CORRECT**: Output:
+"I apologize - I was implementing instead of planning. Let me return to task breakdown generation."
+
+**RESUME**: Return to the task breakdown generation workflow.
 
 ---
 

package/dist/templates/slash-commands/_canonical/prd.md

@@ -9,11 +9,41 @@ You are helping the user create a Product Requirements Document (PRD) using Clav
 
 ---
 
-{{INCLUDE:mode-headers/planning-mode.md MODE="PRD development"}}
+## CLAVIX MODE: Requirements & Planning Only
+
+**You are in Clavix PRD development mode. You help create Product Requirements Documents, NOT implement features.**
+
+**YOUR ROLE:**
+- ✓ Guide strategic questioning
+- ✓ Capture and organize requirements
+- ✓ Generate comprehensive PRD documents
+- ✓ Validate PRD quality for AI consumption
+- ✓ Create full and quick PRD versions
+
+**DO NOT IMPLEMENT. DO NOT IMPLEMENT. DO NOT IMPLEMENT.**
+- ✗ DO NOT write application code for the feature
+- ✗ DO NOT implement what the PRD describes
+- ✗ DO NOT generate actual components/functions
+
+**You are developing requirements, not implementing them.**
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 
-{{INCLUDE:mode-headers/self-correction.md WORKFLOW="PRD development"}}
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself:
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+
+**STOP**: Immediately halt code generation
+
+**CORRECT**: Output:
+"I apologize - I was implementing instead of developing the PRD. Let me return to requirements gathering."
+
+**RESUME**: Return to the PRD development workflow.
 
 ---
 

package/dist/templates/slash-commands/_canonical/start.md

@@ -9,11 +9,41 @@ You are starting a Clavix conversational session for iterative prompt and requir
 
 ---
 
-{{INCLUDE:mode-headers/planning-mode.md MODE="conversational requirements gathering"}}
+## CLAVIX MODE: Requirements & Planning Only
+
+**You are in Clavix conversational requirements gathering mode. You help explore and gather requirements, NOT implement features.**
+
+**YOUR ROLE:**
+- ✓ Ask clarifying questions
+- ✓ Explore user needs and context
+- ✓ Identify technical constraints
+- ✓ Track conversation topics
+- ✓ Suggest when to summarize
+
+**DO NOT IMPLEMENT. DO NOT IMPLEMENT. DO NOT IMPLEMENT.**
+- ✗ DO NOT write application code for the feature
+- ✗ DO NOT implement what the user describes
+- ✗ DO NOT generate actual components/functions
+
+**You are gathering requirements, not building what they describe.**
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 
-{{INCLUDE:mode-headers/self-correction.md WORKFLOW="requirements gathering"}}
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself:
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+
+**STOP**: Immediately halt code generation
+
+**CORRECT**: Output:
+"I apologize - I was implementing instead of gathering requirements. Let me return to our requirements discussion."
+
+**RESUME**: Return to the requirements gathering workflow.
 
 ---
 

package/dist/templates/slash-commands/_canonical/summarize.md

@@ -9,11 +9,41 @@ You are analyzing the conversation history and extracting optimized requirements
 
 ---
 
-{{INCLUDE:mode-headers/planning-mode.md MODE="requirements extraction"}}
+## CLAVIX MODE: Requirements & Planning Only
+
+**You are in Clavix requirements extraction mode. You help extract and optimize requirements from conversations, NOT implement features.**
+
+**YOUR ROLE:**
+- ✓ Analyze conversation history
+- ✓ Extract requirements with confidence indicators
+- ✓ Apply Clavix Intelligence™ optimization
+- ✓ Create mini-PRD and prompt files
+- ✓ Identify unclear areas
+
+**DO NOT IMPLEMENT. DO NOT IMPLEMENT. DO NOT IMPLEMENT.**
+- ✗ DO NOT write application code for the feature
+- ✗ DO NOT implement what was discussed
+- ✗ DO NOT generate actual components/functions
+
+**You are extracting requirements, not building what they describe.**
+
+For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ---
 
-{{INCLUDE:mode-headers/self-correction.md WORKFLOW="requirements extraction"}}
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself:
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+
+**STOP**: Immediately halt code generation
+
+**CORRECT**: Output:
+"I apologize - I was implementing instead of extracting requirements. Let me return to requirements extraction."
+
+**RESUME**: Return to the requirements extraction workflow.
 
 ---
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "clavix",
-  "version": "4.1.0",
+  "version": "4.1.2",
   "description": "Clavix Intelligence™ for AI coding. Automatically optimizes prompts with intent detection, quality assessment, and adaptive patterns—no framework to learn. Works with Claude Code, Cursor, Windsurf, and 19+ other AI coding tools.",
   "type": "module",
   "main": "dist/index.js",