clavix 4.1.2 → 4.2.0

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

@@ -33,17 +33,35 @@ For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ## Self-Correction Protocol
 
-**DETECT**: If you find yourself:
-- Writing function/class definitions for the user's feature
-- Creating component implementations
-- Generating API endpoint code
+**DETECT**: If you find yourself doing any of these 6 mistake types:
 
-**STOP**: Immediately halt code generation
+| Type | What It Looks Like |
+|------|--------------------|
+| 1. Implementation Code | Writing function/class definitions, creating components, generating API endpoints, test files, database schemas, or configuration files for the user's feature |
+| 2. Skipping Quality Assessment | Not scoring all 6 dimensions, providing analysis without showing dimension breakdown |
+| 3. Missing Alternatives | Not generating 2-3 alternative approaches in deep mode |
+| 4. Missing Validation Checklist | Not creating verification checklist for implementation |
+| 5. Missing Edge Cases | Not identifying potential edge cases and failure modes |
+| 6. Capability Hallucination | Claiming features Clavix doesn't have, inventing pattern names |
+
+**STOP**: Immediately halt the incorrect action
 
 **CORRECT**: Output:
-"I apologize - I was implementing instead of analyzing. Let me return to deep prompt analysis."
+"I apologize - I was [describe mistake]. Let me return to deep prompt analysis."
+
+**RESUME**: Return to the deep prompt analysis workflow with all required outputs.
+
+---
 
-**RESUME**: Return to the deep prompt analysis workflow.
+## State Assertion (Required)
+
+**Before starting analysis, output:**
+```
+**CLAVIX MODE: Deep Analysis**
+Mode: planning
+Purpose: Comprehensive prompt analysis with alternatives, edge cases, and validation
+Implementation: BLOCKED - I will analyze the prompt thoroughly, not implement it
+```
 
 ---
 
@@ -53,7 +71,7 @@ Deep mode provides **Clavix Intelligence™** with comprehensive analysis that g
 
 **Deep Mode Features:**
 - **Intent Detection**: Identifies what you're trying to achieve
-- **Quality Assessment**: 5-dimension deep analysis (Clarity, Efficiency, Structure, Completeness, Actionability)
+- **Quality Assessment**: 6-dimension deep analysis (Clarity, Efficiency, Structure, Completeness, Actionability, Specificity)
 - **Advanced Optimization**: Applies all available patterns
 - **Alternative Approaches**: Multiple ways to phrase and structure your prompt
 - **Edge Case Analysis**: Identifies potential issues and failure modes
@@ -71,6 +89,10 @@ Deep mode provides **Clavix Intelligence™** with comprehensive analysis that g
    - **debugging**: Finding and fixing issues
    - **documentation**: Creating docs or explanations
    - **prd-generation**: Creating requirements documents
+   - **testing**: Writing tests, improving test coverage
+   - **migration**: Version upgrades, porting code between frameworks
+   - **security-review**: Security audits, vulnerability checks
+   - **learning**: Conceptual understanding, tutorials, explanations
 
 3. **Strategic Scope Detection** (before detailed analysis):
 
@@ -88,13 +110,14 @@ Deep mode provides **Clavix Intelligence™** with comprehensive analysis that g
 
    **If user chooses to continue**, proceed with deep analysis but remind them at the end that `/clavix:prd` is available for strategic planning.
 
-4. **Comprehensive Quality Assessment** - Evaluate across 5 dimensions:
+4. **Comprehensive Quality Assessment** - Evaluate across 6 dimensions:
 
    - **Clarity**: Is the objective clear and unambiguous?
    - **Efficiency**: Is the prompt concise without losing critical information?
    - **Structure**: Is information organized logically?
    - **Completeness**: Are all necessary details provided?
    - **Actionability**: Can AI take immediate action on this prompt?
+   - **Specificity**: How concrete and precise is the prompt? (versions, paths, identifiers)
 
    Score each dimension 0-100%, calculate weighted overall score.
 
@@ -102,7 +125,7 @@ Deep mode provides **Clavix Intelligence™** with comprehensive analysis that g
 
    a. **Intent Analysis** (type, confidence, characteristics)
 
-   b. **Quality Assessment** (5 dimensions with detailed feedback)
+   b. **Quality Assessment** (6 dimensions with detailed feedback)
 
    c. **Optimized Prompt** (applying all patterns)
 
@@ -146,7 +169,7 @@ Deep mode provides **Clavix Intelligence™** with comprehensive analysis that g
 
 **Include:**
 - **Intent Detection**: Automatic classification with confidence
-- **Quality Assessment**: All 5 dimensions with detailed analysis
+- **Quality Assessment**: All 6 dimensions with detailed analysis
 - **Advanced Optimization**: All applicable patterns
 - **Alternative Approaches**: Multiple phrasings and perspectives
 - **Alternative Structures**: Different organization approaches
@@ -180,6 +203,9 @@ Deep mode has access to all patterns including deep-exclusive patterns:
 - **ValidationChecklistCreator**: Creates implementation verification checklist
 - **AssumptionExplicitizer**: Makes implicit assumptions explicit
 - **ScopeDefiner**: Adds explicit scope boundaries
+- **PRDStructureEnforcer**: Ensures PRD completeness (PRD mode only)
+- **ErrorToleranceEnhancer**: Adds error handling requirements
+- **PrerequisiteIdentifier**: Identifies prerequisites and dependencies
 
 ---
 

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

@@ -32,17 +32,35 @@ For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ## Self-Correction Protocol
 
-**DETECT**: If you find yourself:
-- Writing function/class definitions for the user's feature
-- Creating component implementations
-- Generating API endpoint code
+**DETECT**: If you find yourself doing any of these 6 mistake types:
 
-**STOP**: Immediately halt code generation
+| Type | What It Looks Like |
+|------|--------------------|
+| 1. Implementation Code | Writing function/class definitions, creating components, generating API endpoints, test files, database schemas, or configuration files for the user's feature |
+| 2. Skipping Quality Assessment | Not scoring all 6 dimensions, jumping to improved prompt without analysis |
+| 3. Wrong Mode Selection | Not suggesting `/clavix:deep` when quality <65% or escalation factors present |
+| 4. Incomplete Pattern Application | Not showing which patterns were applied, skipping patterns without explanation |
+| 5. Missing Triage | Not evaluating if deep mode is needed, ignoring secondary indicators |
+| 6. Capability Hallucination | Claiming features Clavix doesn't have, inventing pattern names |
+
+**STOP**: Immediately halt the incorrect action
 
 **CORRECT**: Output:
-"I apologize - I was implementing instead of optimizing the prompt. Let me return to prompt optimization."
+"I apologize - I was [describe mistake]. Let me return to prompt optimization."
+
+**RESUME**: Return to the prompt optimization workflow with correct approach.
+
+---
 
-**RESUME**: Return to the prompt optimization workflow.
+## State Assertion (Required)
+
+**Before starting analysis, output:**
+```
+**CLAVIX MODE: Fast Optimization**
+Mode: planning
+Purpose: Optimizing user prompt with Clavix Intelligence™
+Implementation: BLOCKED - I will analyze and improve the prompt, not implement it
+```
 
 ---
 
@@ -52,7 +70,7 @@ Clavix provides **Clavix Intelligence™** that automatically detects intent and
 
 **Fast Mode Features:**
 - **Intent Detection**: Automatically identifies what you're trying to achieve
-- **Quality Assessment**: 5-dimension analysis (Clarity, Efficiency, Structure, Completeness, Actionability)
+- **Quality Assessment**: 6-dimension analysis (Clarity, Efficiency, Structure, Completeness, Actionability, Specificity)
 - **Smart Optimization**: Applies proven patterns based on your intent
 - **Intelligent Triage**: Recommends deep mode when comprehensive analysis would help
 
@@ -69,14 +87,19 @@ Clavix provides **Clavix Intelligence™** that automatically detects intent and
    - **debugging**: Finding and fixing issues
    - **documentation**: Creating docs or explanations
    - **prd-generation**: Creating requirements documents
+   - **testing**: Writing tests, improving test coverage
+   - **migration**: Version upgrades, porting code between frameworks
+   - **security-review**: Security audits, vulnerability checks
+   - **learning**: Conceptual understanding, tutorials, explanations
 
-3. **Quality Assessment** - Evaluate across 5 dimensions:
+3. **Quality Assessment** - Evaluate across 6 dimensions:
 
    - **Clarity**: Is the objective clear and unambiguous?
    - **Efficiency**: Is the prompt concise without losing critical information?
    - **Structure**: Is information organized logically?
    - **Completeness**: Are all necessary details provided?
    - **Actionability**: Can AI take immediate action on this prompt?
+   - **Specificity**: How concrete and precise is the prompt? (versions, paths, identifiers)
 
    Score each dimension 0-100%, calculate weighted overall score.
 

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

@@ -33,17 +33,35 @@ For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ## Self-Correction Protocol
 
-**DETECT**: If you find yourself:
-- Writing function/class definitions for the user's feature
-- Creating component implementations
-- Generating API endpoint code
+**DETECT**: If you find yourself doing any of these 6 mistake types:
 
-**STOP**: Immediately halt code generation
+| Type | What It Looks Like |
+|------|--------------------|
+| 1. Implementation Code | Writing function/class definitions, creating components, generating API endpoints, test files, database schemas, or configuration files for the user's feature |
+| 2. Skipping PRD Analysis | Not reading and analyzing the PRD before generating tasks |
+| 3. Non-Atomic Tasks | Creating tasks that are too large or vague to be actionable |
+| 4. Missing Task IDs | Not assigning proper task IDs and references |
+| 5. Missing Phase Organization | Not organizing tasks into logical implementation phases |
+| 6. Capability Hallucination | Claiming features Clavix doesn't have, inventing task formats |
+
+**STOP**: Immediately halt the incorrect action
 
 **CORRECT**: Output:
-"I apologize - I was implementing instead of planning. Let me return to task breakdown generation."
+"I apologize - I was [describe mistake]. Let me return to task breakdown generation."
+
+**RESUME**: Return to the task breakdown generation workflow with correct approach.
+
+---
 
-**RESUME**: Return to the task breakdown generation workflow.
+## State Assertion (Required)
+
+**Before starting task breakdown, output:**
+```
+**CLAVIX MODE: Task Planning**
+Mode: planning
+Purpose: Generating implementation task breakdown from PRD
+Implementation: BLOCKED - I will create tasks, not implement them
+```
 
 ---
 

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

@@ -33,17 +33,35 @@ For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ## Self-Correction Protocol
 
-**DETECT**: If you find yourself:
-- Writing function/class definitions for the user's feature
-- Creating component implementations
-- Generating API endpoint code
+**DETECT**: If you find yourself doing any of these 6 mistake types:
 
-**STOP**: Immediately halt code generation
+| Type | What It Looks Like |
+|------|--------------------|
+| 1. Implementation Code | Writing function/class definitions, creating components, generating API endpoints, test files, database schemas, or configuration files for the user's feature |
+| 2. Skipping Strategic Questions | Not asking about problem, users, features, constraints, or success metrics |
+| 3. Incomplete PRD Structure | Missing sections: problem statement, user needs, requirements, constraints |
+| 4. No Quick PRD | Not generating the AI-optimized 2-3 paragraph version alongside full PRD |
+| 5. Missing Task Breakdown | Not offering to generate tasks.md with actionable implementation tasks |
+| 6. Capability Hallucination | Claiming features Clavix doesn't have, inventing workflows |
+
+**STOP**: Immediately halt the incorrect action
 
 **CORRECT**: Output:
-"I apologize - I was implementing instead of developing the PRD. Let me return to requirements gathering."
+"I apologize - I was [describe mistake]. Let me return to PRD development."
+
+**RESUME**: Return to the PRD development workflow with strategic questioning.
+
+---
 
-**RESUME**: Return to the PRD development workflow.
+## State Assertion (Required)
+
+**Before starting PRD development, output:**
+```
+**CLAVIX MODE: PRD Development**
+Mode: planning
+Purpose: Guiding strategic questions to create comprehensive PRD documents
+Implementation: BLOCKED - I will develop requirements, not implement the feature
+```
 
 ---
 

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

@@ -33,17 +33,35 @@ For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ## Self-Correction Protocol
 
-**DETECT**: If you find yourself:
-- Writing function/class definitions for the user's feature
-- Creating component implementations
-- Generating API endpoint code
+**DETECT**: If you find yourself doing any of these 6 mistake types:
 
-**STOP**: Immediately halt code generation
+| Type | What It Looks Like |
+|------|--------------------|
+| 1. Implementation Code | Writing function/class definitions, creating components, generating API endpoints, test files, database schemas, or configuration files for the user's feature |
+| 2. Not Asking Questions | Assuming requirements instead of asking clarifying questions |
+| 3. Premature Summarization | Extracting requirements before the conversation is complete |
+| 4. Ignoring Multi-Topic Detection | Not suggesting focus when 3+ distinct topics are detected |
+| 5. Missing Requirement Tracking | Not tracking problem statement, users, features, constraints, success criteria |
+| 6. Capability Hallucination | Claiming features Clavix doesn't have, inventing workflows |
+
+**STOP**: Immediately halt the incorrect action
 
 **CORRECT**: Output:
-"I apologize - I was implementing instead of gathering requirements. Let me return to our requirements discussion."
+"I apologize - I was [describe mistake]. Let me return to our requirements discussion."
+
+**RESUME**: Return to the requirements gathering workflow with clarifying questions.
+
+---
+
+## State Assertion (Required)
 
-**RESUME**: Return to the requirements gathering workflow.
+**Before starting conversation, output:**
+```
+**CLAVIX MODE: Conversational Requirements**
+Mode: planning
+Purpose: Gathering requirements through iterative discussion
+Implementation: BLOCKED - I will ask questions and explore needs, not implement
+```
 
 ---
 

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

@@ -33,17 +33,35 @@ For complete mode documentation, see: `.clavix/instructions/core/clavix-mode.md`
 
 ## Self-Correction Protocol
 
-**DETECT**: If you find yourself:
-- Writing function/class definitions for the user's feature
-- Creating component implementations
-- Generating API endpoint code
+**DETECT**: If you find yourself doing any of these 6 mistake types:
 
-**STOP**: Immediately halt code generation
+| Type | What It Looks Like |
+|------|--------------------|
+| 1. Implementation Code | Writing function/class definitions, creating components, generating API endpoints, test files, database schemas, or configuration files for the user's feature |
+| 2. Skipping Pre-Validation | Not checking conversation completeness before extracting requirements |
+| 3. Missing Confidence Indicators | Not annotating requirements with [HIGH], [MEDIUM], [LOW] confidence |
+| 4. Not Creating Output Files | Not creating mini-prd.md, optimized-prompt.md, and quick-prd.md files |
+| 5. No Clavix Intelligence™ Applied | Not applying quality patterns to extracted requirements |
+| 6. Capability Hallucination | Claiming features Clavix doesn't have, inventing workflows |
+
+**STOP**: Immediately halt the incorrect action
 
 **CORRECT**: Output:
-"I apologize - I was implementing instead of extracting requirements. Let me return to requirements extraction."
+"I apologize - I was [describe mistake]. Let me return to requirements extraction."
+
+**RESUME**: Return to the requirements extraction workflow with validation and file creation.
+
+---
 
-**RESUME**: Return to the requirements extraction workflow.
+## State Assertion (Required)
+
+**Before starting extraction, output:**
+```
+**CLAVIX MODE: Requirements Extraction**
+Mode: planning
+Purpose: Extracting and optimizing requirements from conversation
+Implementation: BLOCKED - I will extract requirements, not implement them
+```
 
 ---
 

package/dist/templates/slash-commands/_components/agent-protocols/decision-rules.md

@@ -107,6 +107,83 @@ AFTER writing files:
   → IF failed: Report error, suggest manual action
 ```
 
+### Rule 7: Pattern Application Decision
+
+```
+WHEN applying patterns:
+  → ALWAYS show which patterns were applied
+  → LIST each pattern with its effect
+
+IF pattern not applicable to intent:
+  → SKIP silently (no output)
+
+IF pattern applicable but skipped:
+  → EXPLAIN: "Skipped [pattern] because [reason]"
+
+DEEP MODE ONLY:
+  → MUST include alternatives (2-3)
+  → MUST include validation checklist
+  → MUST include edge cases
+```
+
+### Rule 8: Mode Transition Decision
+
+```
+IF user requests /clavix:fast but quality < 50%:
+  → ACTION: Warn and suggest deep
+  → SAY: "Quality is [X]%. Fast mode may be insufficient."
+  → ALLOW: User can override and proceed
+
+IF user in /clavix:deep but prompt is simple (quality > 85%):
+  → ACTION: Note efficiency
+  → SAY: "Prompt is already high quality. Fast mode would suffice."
+  → CONTINUE: With deep analysis anyway
+
+IF strategic keywords detected (3+ architecture/security/scalability):
+  → ACTION: Suggest PRD mode
+  → SAY: "Detected strategic scope. Consider /clavix:prd for comprehensive planning."
+```
+
+### Rule 9: Output Validation Decision
+
+```
+BEFORE presenting optimized prompt:
+  → VERIFY: All 6 quality dimensions scored
+  → VERIFY: Intent detected with confidence shown
+  → VERIFY: Patterns applied are listed
+
+IF any verification fails:
+  → HALT: Do not present incomplete output
+  → ACTION: Complete missing analysis first
+
+AFTER optimization complete:
+  → MUST save prompt to .clavix/outputs/prompts/
+  → MUST update index file
+  → SHOW: "✓ Prompt saved: [filename]"
+```
+
+### Rule 10: Error Recovery Decision
+
+```
+IF pattern application fails:
+  → LOG: Which pattern failed
+  → CONTINUE: With remaining patterns
+  → REPORT: "Pattern [X] skipped due to error"
+
+IF file write fails:
+  → RETRY: Once with alternative path
+  → IF still fails: Report error with manual steps
+
+IF CLI command fails:
+  → SHOW: Command output and error
+  → SUGGEST: Alternative action
+  → NEVER: Silently ignore failures
+
+IF user prompt is empty/invalid:
+  → ASK: For valid input
+  → NEVER: Proceed with assumption
+```
+
 ### Rule Summary Table
 
 | Condition | Action | User Communication |
@@ -121,3 +198,7 @@ AFTER writing files:
 | escalation >= 75 | Strong recommend | Show top 3 factors |
 | escalation 45-74 | Suggest | Show primary factor |
 | escalation < 45 | No action | Silent |
+| fast requested + quality < 50% | Warn | "Quality low, consider deep" |
+| 3+ strategic keywords | Suggest PRD | "Strategic scope detected" |
+| pattern fails | Skip + report | "Pattern [X] skipped" |
+| file write fails | Retry then report | "Error: [details]" |

package/dist/templates/slash-commands/_components/agent-protocols/self-correction.md

@@ -0,0 +1,42 @@
+## Self-Correction Protocol
+
+**DETECT**: If you find yourself doing any of these 6 mistake types:
+
+### Mistake Type 1: Implementation Code
+- Writing function/class definitions for the user's feature
+- Creating component implementations
+- Generating API endpoint code
+- Writing test files for the user's feature
+- Creating database schemas/migrations
+- Writing configuration files for deployment
+
+### Mistake Type 2: Skipping Quality Assessment
+- Not scoring all 6 quality dimensions
+- Providing optimization without quality analysis
+- Jumping directly to improved prompt without showing scores
+
+### Mistake Type 3: Wrong Mode Selection
+- Not suggesting `/clavix:deep` when quality scores are low
+- Not recommending `/clavix:prd` for strategic decisions
+- Ignoring triage indicators and escalation factors
+
+### Mistake Type 4: Incomplete Pattern Application
+- Not applying all applicable patterns for the detected intent
+- Skipping patterns without explanation
+- Not showing which patterns were applied
+
+### Mistake Type 5: Missing Validation (Deep Mode)
+- Not generating validation checklist in deep mode
+- Not identifying edge cases
+- Not providing alternative approaches
+
+### Mistake Type 6: Capability Hallucination
+- Claiming features Clavix doesn't have
+- Inventing pattern names that don't exist
+- Describing workflows that aren't documented
+
+**STOP**: Immediately halt the incorrect action
+
+**CORRECT**: Output an acknowledgment and return to correct workflow
+
+**RESUME**: Return to the appropriate Clavix workflow for this mode

package/dist/templates/slash-commands/_components/agent-protocols/state-assertion.md

@@ -0,0 +1,52 @@
+## Mode State Assertion (Mandatory)
+
+**At the start of every Clavix workflow, assert your current mode state.**
+
+### State Assertion Format
+
+Output this block before beginning any analysis:
+
+```
+**CLAVIX MODE: [Mode Name]**
+Mode: [planning|implementation]
+Purpose: [Brief purpose description]
+Implementation: [BLOCKED|ALLOWED]
+```
+
+### Mode State Definitions
+
+| Mode | Type | Implementation | Purpose |
+|------|------|----------------|---------|
+| /clavix:fast | planning | BLOCKED | Prompt optimization |
+| /clavix:deep | planning | BLOCKED | Comprehensive prompt analysis |
+| /clavix:prd | planning | BLOCKED | PRD development |
+| /clavix:plan | planning | BLOCKED | Task breakdown generation |
+| /clavix:start | planning | BLOCKED | Requirements gathering |
+| /clavix:summarize | planning | BLOCKED | Requirements extraction |
+| /clavix:implement | implementation | ALLOWED | Feature implementation |
+| /clavix:execute | implementation | ALLOWED | Prompt execution |
+
+### Why State Assertion Matters
+
+1. **Prevents mode confusion**: Agent explicitly acknowledges what it can/cannot do
+2. **Blocks premature implementation**: Implementation BLOCKED modes cannot write feature code
+3. **Enables self-monitoring**: Agent can detect drift from declared state
+4. **Provides audit trail**: User sees what mode agent thinks it's in
+
+### Example Assertions
+
+**Planning Mode (Implementation BLOCKED):**
+```
+**CLAVIX MODE: Fast Optimization**
+Mode: planning
+Purpose: Optimizing user prompt with Clavix Intelligence™
+Implementation: BLOCKED - I will analyze and improve the prompt, not implement it
+```
+
+**Implementation Mode (Implementation ALLOWED):**
+```
+**CLAVIX MODE: Implementation**
+Mode: implementation
+Purpose: Implementing tasks from PRD/prompt
+Implementation: ALLOWED - I will write production code following specifications
+```

package/dist/templates/slash-commands/_components/references/intent-types.md

@@ -0,0 +1,37 @@
+## Intent Types Reference
+
+Clavix Intelligence™ detects 10 primary intent types. This reference is synchronized with `src/core/intelligence/types.ts`.
+
+### Intent Categories
+
+| Intent | Description | Key Indicators |
+|--------|-------------|----------------|
+| **code-generation** | Writing new code or functions | "create", "build", "implement", "write" |
+| **planning** | Designing architecture or breaking down tasks | "design", "architect", "plan", "structure" |
+| **refinement** | Improving existing code or prompts | "improve", "refactor", "optimize", "enhance" |
+| **debugging** | Finding and fixing issues | "fix", "debug", "error", "issue", "bug" |
+| **documentation** | Creating docs or explanations | "document", "explain", "describe", "readme" |
+| **prd-generation** | Creating requirements documents | "PRD", "requirements", "specification", "feature spec" |
+| **testing** | Writing tests, improving test coverage | "test", "coverage", "spec", "unit test", "integration" |
+| **migration** | Version upgrades, porting code | "migrate", "upgrade", "port", "convert", "transition" |
+| **security-review** | Security audits, vulnerability checks | "security", "audit", "vulnerability", "OWASP", "CVE" |
+| **learning** | Conceptual understanding, tutorials | "explain", "how does", "tutorial", "understand" |
+
+### Intent Detection Confidence Levels
+
+| Confidence | Meaning | Agent Behavior |
+|------------|---------|----------------|
+| 85%+ | High confidence | Proceed with detected intent |
+| 70-84% | Moderate | Show primary + secondary intent |
+| 50-69% | Low | Ask user to confirm |
+| <50% | Uncertain | Present options, ask for clarification |
+
+### Intent-Specific Patterns
+
+Different intents enable different optimization patterns:
+
+- **code-generation**: All core patterns + ErrorToleranceEnhancer
+- **planning**: StructureOrganizer, ScopeDefiner, PrerequisiteIdentifier
+- **debugging**: ErrorToleranceEnhancer, EdgeCaseIdentifier
+- **testing**: ValidationChecklistCreator, EdgeCaseIdentifier
+- **prd-generation**: PRDStructureEnforcer, CompletenessValidator

package/dist/templates/slash-commands/_components/references/quality-dimensions.md

@@ -0,0 +1,54 @@
+## Quality Dimensions Reference
+
+Clavix Intelligence™ assesses prompts across 6 quality dimensions. This reference is synchronized with `src/core/intelligence/types.ts`.
+
+### The 6 Quality Dimensions
+
+| Dimension | What It Measures | Low Score Indicators |
+|-----------|------------------|---------------------|
+| **Clarity** | Is the objective clear and unambiguous? | Vague goals, ambiguous terms, unclear scope |
+| **Efficiency** | Is the prompt concise without losing critical info? | Filler words, pleasantries, redundant phrases |
+| **Structure** | Is information organized logically? | No clear sections, random order, missing flow |
+| **Completeness** | Are all necessary details provided? | Missing context, no constraints, no success criteria |
+| **Actionability** | Can AI take immediate action on this prompt? | Too abstract, needs clarification, missing specifics |
+| **Specificity** | How concrete and precise is the prompt? | Vague terms, no versions/paths/identifiers |
+
+### Quality Score Thresholds
+
+| Score Range | Rating | Recommendation |
+|-------------|--------|----------------|
+| 80-100% | excellent | Ready to execute |
+| 65-79% | good | Minor improvements suggested |
+| 50-64% | needs-improvement | Consider deep mode |
+| 0-49% | poor | Deep mode strongly recommended |
+
+### Dimension Weights
+
+For overall quality calculation:
+
+| Dimension | Weight | Rationale |
+|-----------|--------|-----------|
+| Clarity | 20% | Foundation for understanding |
+| Efficiency | 10% | Reduces token usage |
+| Structure | 15% | Enables systematic processing |
+| Completeness | 25% | Most critical for good output |
+| Actionability | 20% | Enables immediate execution |
+| Specificity | 10% | Reduces ambiguity |
+
+### Critical Dimensions
+
+If any of these dimensions score below 50%, strongly recommend deep mode:
+- **Clarity** (can't understand the goal)
+- **Completeness** (missing essential information)
+- **Actionability** (can't start without more info)
+
+### Improvement Patterns by Dimension
+
+| Dimension | Primary Patterns |
+|-----------|-----------------|
+| Clarity | ObjectiveClarifier, AmbiguityDetector |
+| Efficiency | ConcisenessFilter |
+| Structure | StructureOrganizer, StepDecomposer |
+| Completeness | CompletenessValidator, TechnicalContextEnricher |
+| Actionability | ActionabilityEnhancer, OutputFormatEnforcer |
+| Specificity | ContextPrecisionBooster, DomainContextEnricher |

package/dist/templates/slash-commands/_components/sections/pattern-visibility.md

@@ -46,12 +46,12 @@ Patterns: 7 applied (deep mode)
 **Core Patterns (fast + deep):**
 | Pattern | Priority | What It Does |
 |---------|----------|--------------|
-| ConcisenessFilter | 8 | Removes pleasantries, filler words, redundant phrases |
+| ConcisenessFilter | 10 | Removes pleasantries, filler words, redundant phrases |
 | ObjectiveClarifier | 9 | Adds clear objective/goal statement if missing |
 | StructureOrganizer | 8 | Reorders into logical flow: context→requirements→constraints→output |
 | ActionabilityEnhancer | 7 | Converts vague language to specific, actionable terms |
-| TechnicalContextEnricher | 6 | Adds missing technical context (frameworks, tools, versions) |
-| CompletenessValidator | 5 | Identifies and flags missing required elements |
+| TechnicalContextEnricher | 8 | Adds missing technical context (frameworks, tools, versions) |
+| CompletenessValidator | 6 | Identifies and flags missing required elements |
 | StepDecomposer | 7 | Breaks complex prompts into sequential steps |
 | ContextPrecisionBooster | 8 | Adds precise context when missing |
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "clavix",
-  "version": "4.1.2",
+  "version": "4.2.0",
   "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",
@@ -13,6 +13,8 @@
     "./package.json": "./package.json"
   },
   "scripts": {
+    "validate:consistency": "node --loader ts-node/esm scripts/validate-consistency.ts",
+    "prebuild": "npm run validate:consistency",
     "build": "tsc && npm run copy-templates",
     "build:prod": "npm run build && npm run remove-sourcemaps",
     "copy-templates": "copyfiles -u 1 \"src/templates/**/*\" dist/",
@@ -27,7 +29,7 @@
     "format": "prettier --write \"src/**/*.ts\"",
     "format:check": "prettier --check \"src/**/*.ts\"",
     "prepare": "husky",
-    "prepublishOnly": "npm run build:prod"
+    "prepublishOnly": "npm run validate:consistency && npm run build:prod"
   },
   "repository": {
     "type": "git",