clavix 5.10.0 → 5.10.2

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

@@ -274,7 +274,7 @@ Result: Project permanently deleted
 
 ---
 
-## Agent Transparency (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

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

@@ -22,6 +22,21 @@ When you run `/clavix:implement`, I:
 
 **You just say "implement" and I handle the rest.**
 
+### Command Variations
+
+**Parse the slash command content to determine implementation scope:**
+
+| Command Pattern | Interpretation | Action |
+|----------------|----------------|--------|
+| `/clavix:implement` or `/clavix:implement all` | All tasks | Implement all pending tasks |
+| `/clavix:implement task 3` | Single task | Implement only task #3 |
+| `/clavix:implement` (no qualifier) | Unspecified | Ask user to choose |
+
+**How to parse:**
+1. Check if command contains "task <number>" pattern
+2. Check if command contains "all" keyword
+3. If neither → ask user for selection
+
 ### Detection Priority
 
 ```
@@ -59,12 +74,42 @@ Found tasks.md with [N] pending tasks in summarize/ (legacy location). Starting
 
 This confirmation ensures the user knows exactly what will be implemented before any code is written.
 
-### Explicit Flags
+### Task Selection (REQUIRED)
+
+**When Task Implementation Mode is detected, you MUST determine task scope BEFORE starting:**
+
+**Step 1: Parse the slash command**
+- Check command content for: `task <N>` pattern (e.g., "task 3", "task 5")
+- Check command content for: `all` keyword
+- If neither found → proceed to Step 2
 
-Override auto-detection when needed:
-- `--tasks` - Force task mode (skip prompt check)
+**Step 2: If no qualifier in command, ASK the user:**
+> "I found [N] pending tasks in [project-name]. How would you like to proceed?
+>
+> Options:
+> - **all** - Implement all pending tasks
+> - **task <N>** - Implement only task number N (e.g., "task 3")
+> - **list** - Show all tasks with numbers
+>
+> Which would you prefer?"
+
+**Step 3: Handle selection**
+- If "all" in command or user response → Implement all pending tasks
+- If "task N" detected → Validate N exists, implement only that task
+- If "list" requested → Show numbered list of incomplete tasks, ask again
+
+**Step 4: Confirm before starting**
+> "Found tasks.md with [N] pending tasks in [project-name].
+>
+> Mode: [ALL tasks | Single task #N: [task description]]
+> Starting task implementation..."
+
+### Prompt Execution Flags
+
+For prompt execution mode (when tasks.md not found):
 - `--prompt <id>` - Execute specific prompt by ID
 - `--latest` - Execute most recent prompt
+- `--tasks` - Force task mode (skip prompt check, use tasks.md)
 
 ---
 
@@ -103,6 +148,8 @@ If you catch yourself doing any of these, STOP and correct:
 4. **Batch Task Completion** - Marking multiple tasks done without implementing each
 5. **Ignoring Blocked Tasks** - Not reporting when a task cannot be completed
 6. **Capability Hallucination** - Claiming Clavix can do things it cannot
+7. **Not Parsing Command** - Not checking command content for "task N" or "all" before asking
+8. **Wrong Task Number** - Not validating task number is within range before implementing
 
 **DETECT → STOP → CORRECT → RESUME**
 
@@ -293,32 +340,54 @@ At the end of our session, I'll remind you:
 
 ## Example: What a Session Looks Like
 
+### Single Task Mode
+
 ```
-You:    /clavix:implement
+You:    /clavix:implement task 3
 
 Me:     "Found your task list! Let me see where we are..."
 
-        "📋 Progress: 0/8 tasks complete
+        "📋 8 total tasks, 0 complete
 
-        Starting with: Set up project structure (phase-1-setup-1)
+        Implementing task 3: Create database models (phase-1-database-1)
 
-        Let me build this..."
+        I see this needs PostgreSQL with Prisma. Let me check your .env..."
 
-        [I create folders, initial files, configure package.json]
+        [I create Prisma schema, run migrations]
 
-        "✓ Done! Setting up next task..."
+        "✓ Task 3 complete!
 
-        "Working on: Create database models (phase-1-database-1)
+        Progress: 1/8 tasks (12.5%)
+        Checkbox updated in tasks.md
 
-        I see this needs PostgreSQL with Prisma. Let me check your .env..."
+        Would you like to continue with another task?"
+```
 
-        [I create Prisma schema, run migrations]
+### All Tasks Mode (with user prompt)
 
-        "✓ Done! 2/8 tasks complete (25%)
+```
+You:    /clavix:implement
+
+Me:     "Found your task list! Let me see where we are..."
+
+        "📋 8 total tasks, 0 complete
+
+        How would you like to proceed?
+        - **all** - Implement all pending tasks
+        - **task <N>** - Implement only task number N
+        - **list** - Show all tasks with numbers"
+
+You:    all
+
+Me:     "Implementing all 8 pending tasks...
+
+        Starting with task 1: Set up project structure (phase-1-setup-1)
+
+        [I create folders, initial files, configure package.json]
 
-        Next: Set up authentication..."
+        ✓ Done! Moving to task 2..."
 
-        [... continues through all tasks ...]
+        [continues through all tasks]
 
 Me:     "🎉 All 8 tasks complete!
 
@@ -331,6 +400,28 @@ Me:     "🎉 All 8 tasks complete!
         Ready to test it out?"
 ```
 
+### List Mode
+
+```
+You:    /clavix:implement
+
+Me:     "I found 8 pending tasks. How would you like to proceed?
+        - **all** - Implement all pending tasks
+        - **task <N>** - Implement only task number N
+        - **list** - Show all tasks with numbers"
+
+You:    list
+
+Me:     "Here are all pending tasks:
+        1. Set up project structure (phase-1-setup-1)
+        2. Create database models (phase-1-setup-2)
+        3. Set up authentication (phase-2-auth-1)
+        4. Build user registration (phase-2-auth-2)
+        ...
+
+        Which task would you like to implement?"
+```
+
 ## How I Find Tasks
 
 Task IDs look like: `phase-1-setup-1`, `phase-2-auth-3`
@@ -524,7 +615,7 @@ I'll explain what's wrong and what you might need to do:
 
 ---
 
-## Agent Transparency (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### 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 (v5.10.0)
+## Agent Transparency (v5.10.2)
 
 ### Agent Manual (Universal Protocols)
 {{INCLUDE:agent-protocols/AGENT_MANUAL.md}}

package/dist/utils/path-resolver.js

@@ -7,6 +7,7 @@
  * 3. Built-in default from integrations.json
  */
 import * as os from 'os';
+import * as path from 'path';
 /**
  * Mapping of integration names to their environment variable names
  *
@@ -29,18 +30,35 @@ export const ENV_VAR_MAP = {
  */
 export function resolveIntegrationPath(config, userConfig) {
     const integrationName = config.name;
+    let resolvedPath;
     // Priority 1: Environment variable
     const envVar = ENV_VAR_MAP[integrationName];
     if (envVar && process.env[envVar]) {
-        return process.env[envVar];
+        resolvedPath = process.env[envVar];
     }
-    // Priority 2: User config override
-    const customPath = userConfig?.experimental?.integrationPaths?.[integrationName];
-    if (customPath && typeof customPath === 'string') {
-        return expandTilde(customPath);
+    else {
+        // Priority 2: User config override
+        const customPath = userConfig?.experimental?.integrationPaths?.[integrationName];
+        if (customPath && typeof customPath === 'string') {
+            resolvedPath = customPath;
+        }
+        else {
+            // Priority 3: Built-in default from config
+            resolvedPath = config.directory;
+        }
     }
-    // Priority 3: Built-in default from config
-    return expandTilde(config.directory);
+    // Codex-specific: ensure /prompts subdirectory exists
+    // When using $CODEX_HOME, the user may set it to the base directory
+    // (e.g., /custom/codex) without the /prompts suffix. We need to
+    // ensure prompts always go into the /prompts subdirectory.
+    if (integrationName === 'codex') {
+        const normalizedPath = path.normalize(resolvedPath);
+        if (!normalizedPath.endsWith(path.join('prompts')) &&
+            !normalizedPath.endsWith(path.sep + 'prompts')) {
+            resolvedPath = path.join(resolvedPath, 'prompts');
+        }
+    }
+    return expandTilde(resolvedPath);
 }
 /**
  * Expands tilde (~) to the user's home directory

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "clavix",
-  "version": "5.10.0",
+  "version": "5.10.2",
   "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:archive    Archive completed projects\n\nWorks with Claude Code, Cursor, Windsurf, and 20 AI coding tools.",
   "type": "module",
   "main": "dist/index.js",