clavix 7.2.0 → 7.2.2

package/dist/templates/skills/prd.md

@@ -1,9 +1,8 @@
 ---
 name: clavix-prd
-description: Create comprehensive PRDs through strategic questioning
-license: MIT
+description: Create comprehensive Product Requirements Documents through strategic questioning. Use when planning a new feature or project that needs clear requirements.
+license: Apache-2.0
 ---
-
 # Clavix PRD Skill
 
 Transform ideas into structured Product Requirements Documents through strategic questioning.
@@ -292,10 +291,27 @@ Display quality scores and improvement suggestions if needed.
 
 After PRD creation, guide user to:
 
+**REQUIRED SUB-SKILL**: Use `clavix-plan` to generate implementation tasks
+
+Do not skip directly to implementation. Follow the skill chain:
+
+```
+clavix-prd (you are here)
+      │
+      ▼ REQUIRED
+clavix-plan
+      │
+      ▼ REQUIRED
+clavix-implement
+      │
+      ▼ REQUIRED (after all tasks)
+clavix-verify
+```
+
 | If... | Recommend |
 |-------|-----------|
-| Ready for task breakdown | `/clavix-plan` - Generate tasks.md from PRD |
-| Simple enough to implement directly | `/clavix-implement` |
+| Ready for task breakdown | `/clavix-plan` - Generate tasks.md from PRD (RECOMMENDED) |
+| Simple enough to implement directly | `/clavix-implement` (but still verify after) |
 | Want to improve a prompt first | `/clavix-improve` |
 
 ---
@@ -342,4 +358,4 @@ After PRD creation, guide user to:
 **Solution**:
 - Review each section with user
 - Ask "What's missing or inaccurate?"
-- Update PRD manually or regenerate with corrected answers
+- Update PRD manually or regenerate with corrected answers

package/dist/templates/skills/using-clavix.md

@@ -0,0 +1,269 @@
+---
+name: using-clavix
+description: Use when starting any conversation involving Clavix workflows - establishes skill invocation rules, verification requirements, and workflow orchestration
+license: Apache-2.0
+---
+# Using Clavix
+
+If you think there is even a 1% chance a Clavix skill might apply to what you are doing, you ABSOLUTELY MUST invoke the skill.
+
+IF A CLAVIX SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.
+
+This is not negotiable. This is not optional. You cannot rationalize your way out of this.
+
+---
+
+## How to Access Skills
+
+**In Claude Code / Amp:** Use the `Skill` tool. When you invoke a skill, its content is loaded and presented to you—follow it directly.
+
+**In other environments:** Check your platform's documentation for how skills are loaded.
+
+---
+
+## The Rule
+
+**Invoke relevant Clavix skills BEFORE any response or action.** Even a 1% chance a skill might apply means you should invoke the skill to check. If an invoked skill turns out to be wrong for the situation, you don't need to use it.
+
+```
+User message received
+        │
+        ▼
+"Might any Clavix skill apply?" ──── definitely not ──► Respond normally
+        │
+        │ yes, even 1%
+        ▼
+Invoke Skill tool
+        │
+        ▼
+Announce: "Using [skill] to [purpose]"
+        │
+        ▼
+Follow skill exactly
+        │
+        ▼
+Respond (including clarifications)
+```
+
+---
+
+## Red Flags - STOP and Check
+
+These thoughts mean STOP—you're rationalizing:
+
+| Thought | Reality |
+|---------|---------|
+| "This is just a simple question" | Questions are tasks. Check for skills. |
+| "I need more context first" | Skill check comes BEFORE clarifying questions. |
+| "Let me explore the codebase first" | Skills tell you HOW to explore. Check first. |
+| "I can check files quickly" | Files lack conversation context. Check for skills. |
+| "Let me gather information first" | Skills tell you HOW to gather information. |
+| "This doesn't need a formal skill" | If a skill exists, use it. |
+| "I remember this skill" | Skills evolve. Read current version. |
+| "This doesn't count as a task" | Action = task. Check for skills. |
+| "The skill is overkill" | Simple things become complex. Use it. |
+| "I'll just do this one thing first" | Check BEFORE doing anything. |
+| "I know what that means" | Knowing the concept ≠ using the skill. Invoke it. |
+
+---
+
+## Clavix Skill Priority
+
+When multiple skills could apply, use this order:
+
+1. **Exploration skills first** (clavix-start) - for vague ideas needing discovery
+2. **Planning skills second** (clavix-prd, clavix-plan) - for structuring requirements
+3. **Implementation skills third** (clavix-implement) - for executing tasks
+4. **Verification skills always** (clavix-verify) - NEVER skip after implementation
+
+"Let's build X" → clavix-start or clavix-prd first, then clavix-plan, then clavix-implement.
+"Fix this bug" → Use systematic-debugging, then clavix-implement if needed.
+
+---
+
+## Clavix Workflow Map
+
+```
+[Vague Idea]
+      │
+      ▼
+clavix-start ──► [Conversation] ──► clavix-summarize
+      │                                    │
+      │                                    ▼
+      │                             [Mini-PRD/Prompt]
+      │                                    │
+      ▼                                    ▼
+clavix-prd ◄───────────────────────────────┘
+      │
+      ▼
+[Full PRD + Quick PRD]
+      │
+      ▼
+clavix-plan
+      │
+      ▼
+[tasks.md with Implementation Plan]
+      │
+      ▼
+clavix-implement ◄──────────────────┐
+      │                              │
+      ▼                              │
+[Code Changes]                       │
+      │                              │
+      ▼                              │
+clavix-verify                        │
+      │                              │
+      ├── Issues found? ────────────►│
+      │                     Fix Loop │
+      ▼                              │
+[All Verified] ◄─────────────────────┘
+      │
+      ▼
+clavix-archive (optional)
+```
+
+---
+
+## Required Skill Chains
+
+These chains are MANDATORY. Do not skip steps.
+
+### Planning → Implementation Chain
+
+```
+clavix-prd
+    │
+    ▼ REQUIRED
+clavix-plan
+    │
+    ▼ REQUIRED
+clavix-implement
+    │
+    ▼ REQUIRED
+clavix-verify
+```
+
+### Implementation → Verification Chain
+
+```
+clavix-implement (each task)
+    │
+    ▼ REQUIRED (after ALL tasks complete)
+clavix-verify
+    │
+    ├── Issues found?
+    │       │
+    │       ▼
+    │   Fix the issues
+    │       │
+    │       ▼
+    │   Re-run clavix-verify (REQUIRED)
+    │       │
+    │       └── Repeat until all issues resolved
+    │
+    ▼ (only when all pass)
+Done / clavix-archive
+```
+
+---
+
+## The Iron Laws
+
+### Iron Law 1: No Completion Without Verification
+
+```
+NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE
+```
+
+If you haven't run the verification command in this message, you cannot claim it passes.
+
+### Iron Law 2: No Implementation Without Planning
+
+```
+NO CODE CHANGES WITHOUT A PLAN OR TASK SPECIFICATION
+```
+
+If you're implementing features, there should be a PRD or task list guiding the work.
+
+### Iron Law 3: No Skipping Fix Loops
+
+```
+ISSUES FOUND = ISSUES FIXED + RE-VERIFIED
+```
+
+If verification found issues, you must fix them AND re-verify. Proceeding without re-verification is forbidden.
+
+---
+
+## Verification Gate Pattern
+
+**BEFORE claiming any status or expressing satisfaction:**
+
+1. **IDENTIFY**: What command proves this claim?
+2. **RUN**: Execute the FULL command (fresh, complete)
+3. **READ**: Full output, check exit code, count failures
+4. **VERIFY**: Does output confirm the claim?
+   - If NO: State actual status with evidence
+   - If YES: State claim WITH evidence
+
+**Skip any step = lying, not verifying**
+
+---
+
+## Common Verification Requirements
+
+| Claim | Requires | Not Sufficient |
+|-------|----------|----------------|
+| Tests pass | Test command output: 0 failures | Previous run, "should pass" |
+| Linter clean | Linter output: 0 errors | Partial check, extrapolation |
+| Build succeeds | Build command: exit 0 | Linter passing, logs look good |
+| Bug fixed | Test original symptom: passes | Code changed, assumed fixed |
+| Task complete | Verification evidence shown | "I implemented it" |
+| Requirements met | Line-by-line checklist verified | Tests passing alone |
+
+---
+
+## Clavix Skill Reference
+
+| Skill | When to Use |
+|-------|-------------|
+| `clavix-start` | Ideas are vague, need conversational exploration |
+| `clavix-summarize` | Extract requirements from conversation into mini-PRD |
+| `clavix-prd` | Create comprehensive PRD through strategic questions |
+| `clavix-plan` | Transform PRD into actionable task breakdown |
+| `clavix-implement` | Execute tasks from plan with progress tracking |
+| `clavix-verify` | Audit implementation against PRD requirements |
+| `clavix-review` | Review PR/code changes with criteria analysis |
+| `clavix-refine` | Iterate on existing PRD or prompt |
+| `clavix-improve` | Optimize a prompt with quality assessment |
+| `clavix-archive` | Archive completed project outputs |
+
+---
+
+## Skill Types
+
+**Rigid Skills** (clavix-implement, clavix-verify): Follow exactly. Don't adapt away from the discipline.
+
+**Flexible Skills** (clavix-start, clavix-prd): Adapt principles to context, but maintain the core flow.
+
+The skill itself tells you which type it is through its structure.
+
+---
+
+## User Instructions
+
+Instructions say WHAT, not HOW. "Add X" or "Fix Y" doesn't mean skip workflows.
+
+"Implement this feature" → Check for clavix-prd or clavix-plan first
+"Just make it work" → Still requires verification after implementation
+"Quick fix" → Still requires evidence before claiming success
+
+---
+
+## Remember
+
+- Check for skills BEFORE any action
+- Follow skill chains completely
+- Never skip verification steps
+- Issues found = must fix AND re-verify
+- Evidence before claims, always

package/dist/templates/skills/verify.md

@@ -1,19 +1,33 @@
 ---
 name: clavix-verify
-description: Spec-driven technical audit against requirements and implementation plan
-license: MIT
+description: Verify implementation against PRD requirements with systematic checking. Use after implementation to validate completeness.
+license: Apache-2.0
 ---
-
 # Clavix Verify Skill
 
 Perform a **Spec-Driven Technical Audit** of your implementation. I don't just "run tests"—I verify that your code matches the **Plan** (`tasks.md`) and the **Requirements** (`full-prd.md`).
 
+---
+
+## The Iron Law
+
+```
+ISSUES FOUND = ISSUES FIXED + RE-VERIFIED
+```
+
+If verification found issues, you must fix them AND re-verify. Proceeding without re-verification is forbidden.
+
+**Evidence before claims, always.**
+
+---
+
 ## What This Skill Does
 
 1. **Load the Spec** - Read `full-prd.md` and `tasks.md` to understand requirements and design
 2. **Read the Code** - Inspect actual source files for completed tasks
 3. **Compare & Analyze** - Check implementation accuracy, requirements coverage, code quality
 4. **Generate Review Comments** - Output structured issues (Critical, Major, Minor)
+5. **Enforce Fix Loop** - Issues must be fixed and re-verified before proceeding
 
 ---
 
@@ -176,9 +190,35 @@ Are there quality issues?
 
 ---
 
-## Fixing Workflow (Optional Loop)
+## Fixing Workflow (MANDATORY Loop)
+
+When issues are found, they MUST be fixed. This is not optional.
 
-When user says "Fix #1" or "Fix all critical":
+### The Fix Loop
+
+```
+Issues Found in Report
+         │
+         ▼
+User says "Fix #X" or "Fix all"
+         │
+         ▼
+   Implement Fix
+         │
+         ▼
+   Re-Verify (REQUIRED) ◄───┐
+         │                   │
+         ├── Still issues? ──┘
+         │
+         ▼ (resolved)
+   Issue Marked Resolved
+         │
+         ▼
+   More issues? → Repeat
+         │
+         ▼ (all resolved)
+   Verification Complete
+```
 
 ### Step 1: Acknowledge
 
@@ -188,7 +228,7 @@ When user says "Fix #1" or "Fix all critical":
 
 Modify the code to resolve the specific issue.
 
-### Step 3: Re-Verify
+### Step 3: Re-Verify (REQUIRED)
 
 Run a focused verification on just that file/issue:
 
@@ -196,6 +236,17 @@ Run a focused verification on just that file/issue:
 > - ✅ #1 now uses `apiClient` singleton
 > - Issue resolved"
 
+**If the fix didn't work**, repeat steps 2-3. Do not proceed to next issue.
+
+### Step 4: Evidence
+
+When claiming an issue is fixed:
+- Show the code change made
+- Show the verification output
+- Reference specific file:line
+
+**"Fixed" without evidence = not fixed.**
+
 ---
 
 ## Save Location
@@ -284,13 +335,13 @@ To verify implementation:
 **You are here:** Verify (auditing implementation)
 
 **Common flows:**
-- After implement → `/clavix:verify` → fix issues → verify again
-- Before merge → `/clavix:verify` → ensure quality
+- After implement → `/clavix-verify` → fix issues → verify again
+- Before merge → `/clavix-verify` → ensure quality
 
 **Related commands:**
-- `/clavix:implement` - Build features (what you're verifying)
-- `/clavix:plan` - See the tasks being checked
-- `/clavix:refine` - Update PRD if requirements were wrong
+- `/clavix-implement` - Build features (what you're verifying)
+- `/clavix-plan` - See the tasks being checked
+- `/clavix-refine` - Update PRD if requirements were wrong
 
 ---
 
@@ -298,13 +349,46 @@ To verify implementation:
 
 **If all passed:**
 - ✅ Ready for next phase or merge
-- Consider `/clavix:archive` for completed projects
+- Consider `/clavix-archive` for completed projects
+
+**If failures exist (MANDATORY FIX LOOP):**
+
+```
+Failures in report
+       │
+       ▼
+Fix critical issues first
+       │
+       ▼
+Re-run verification ◄───┐
+       │                 │
+       ├── Still fails ──┘
+       │
+       ▼ (all pass)
+Done
+```
 
-**If failures exist:**
 - Address critical issues first
 - Re-run verification after fixes
-- Repeat until passing
+- **Repeat until passing** - Do NOT skip this
+- Cannot claim "done" until all issues resolved
 
 **If OUTDATED issues:**
 - Update `tasks.md` or `full-prd.md` to reflect reality
 - Re-verify to clear the outdated flags
+
+---
+
+## Red Flags - STOP
+
+If you catch yourself thinking:
+
+| Thought | Reality |
+|---------|---------|
+| "It's probably fixed now" | RUN the verification |
+| "Just this one issue can wait" | Fix it now |
+| "I'll re-verify later" | Re-verify NOW |
+| "The important ones are fixed" | ALL issues must be resolved |
+| "Close enough" | Not done until verified |
+
+**All of these mean: STOP. Follow the fix loop.**

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

@@ -274,7 +274,7 @@ Result: Project permanently deleted
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -615,7 +615,7 @@ I'll explain what's wrong and what you might need to do:
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -524,7 +524,7 @@ Wait for the user to decide what to do next.
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -229,7 +229,7 @@ Present the plan and ask:
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -354,7 +354,7 @@ The validation ensures generated PRDs are immediately usable for AI consumption
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -415,7 +415,7 @@ I'll update the PRD and add this to the refinement history. Confirm?
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -447,7 +447,7 @@ consistent with the project's conventions.
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -230,7 +230,7 @@ The goal is natural exploration of requirements, not a rigid questionnaire. Foll
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -409,7 +409,7 @@ The `/clavix:summarize` command extracts requirements from exploratory conversat
 
 ---
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -123,7 +123,7 @@ Implementation: BLOCKED - I'll analyze and report, not modify or fix
 
 ----
 
-## Agent Transparency (v7.2.0)
+## Agent Transparency (v7.2.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "clavix",
-  "version": "7.2.0",
+  "version": "7.2.2",
   "description": "Agentic-first prompt workflows. Markdown templates that teach AI agents how to optimize prompts, create PRDs, and manage implementation.\n\nSLASH COMMANDS (in your AI assistant):\n  /clavix:improve    Optimize prompts with auto-depth\n  /clavix:prd        Generate PRD through questions\n  /clavix:plan       Create task breakdown from PRD\n  /clavix:implement  Execute tasks with progress tracking\n  /clavix:start      Begin conversational session\n  /clavix:summarize  Extract requirements from conversation\n  /clavix:refine     Refine existing PRD or prompt\n  /clavix:verify     Verify implementation against requirements\n  /clavix:review     Review teammate PRs with criteria\n  /clavix:archive    Archive completed projects\n\nWorks with Claude Code, Cursor, Windsurf, and 20 AI coding tools.",
   "type": "module",
   "main": "dist/index.js",