Strengthen skill delegation in all 11 delegating commands

Replace weak "Invoke the groundwork:X skill" instruction with
CRITICAL INSTRUCTION block that requires Skill tool call as the
first action. Also add Skill to allowed-tools where it was missing.

Author: etr

commands/build-unplanned.md

@@ -6,4 +6,7 @@ allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Task", "Skill"
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:build-unplanned-feature skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:build-unplanned-feature")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/check-specs-alignment.md

@@ -2,8 +2,11 @@
 name: groundwork:check-specs-alignment
 description: Verify alignment between code and specs. Usage /groundwork:check-specs-alignment
 argument-hint: "[alignment-context]"
-allowed-tools: ["Read", "Glob", "Grep", "Task"]
+allowed-tools: ["Read", "Glob", "Grep", "Task", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:check-alignment skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:check-alignment")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/create-tasks.md

@@ -2,8 +2,11 @@
 name: groundwork:create-tasks
 description: Generate implementation tasks from architecture document. Usage /groundwork:create-tasks
 argument-hint: "[filter]"
-allowed-tools: ["Read", "Edit", "Write", "Glob", "Grep", "Task"]
+allowed-tools: ["Read", "Edit", "Write", "Glob", "Grep", "Task", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:tasks skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:tasks")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/design-architecture.md

@@ -2,8 +2,11 @@
 name: groundwork:design-architecture
 description: Create architecture design document with decisions and tradeoffs. Usage /groundwork:design-architecture
 argument-hint: "[feature-name]"
-allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Task"]
+allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Task", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:architecture skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:architecture")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/design-product.md

@@ -2,8 +2,11 @@
 name: groundwork:design-product
 description: Create detailed product requirements document (PRD) for a feature. Usage /groundwork:design-product
 argument-hint: "[product-name]"
-allowed-tools: ["Read", "Edit", "Write", "Glob", "Grep", "Task"]
+allowed-tools: ["Read", "Edit", "Write", "Glob", "Grep", "Task", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:product-design skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:product-design")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/source-architecture-from-code.md

@@ -2,8 +2,11 @@
 name: groundwork:source-architecture-from-code
 description: Synchronize architecture document with codebase changes. Usage /groundwork:source-architecture-from-code
 argument-hint: "[files...]"
-allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep"]
+allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:sync-architecture skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:sync-architecture")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/source-product-specs-from-code.md

@@ -2,8 +2,11 @@
 name: groundwork:source-product-specs-from-code
 description: Synchronize PRD with implemented codebase changes. Usage /groundwork:source-product-specs-from-code
 argument-hint: "[files...]"
-allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep"]
+allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:sync-specs skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:sync-specs")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/source-ux-design-from-code.md

@@ -2,8 +2,11 @@
 name: groundwork:source-ux-design-from-code
 description: Synchronize design system document with codebase changes. Usage /groundwork:source-ux-design-from-code
 argument-hint: "[files...]"
-allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep"]
+allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:sync-design-system skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:sync-design-system")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/ux-design.md

@@ -2,8 +2,11 @@
 name: groundwork:ux-design
 description: Create design system with foundations, brand identity, and UX patterns. Usage /groundwork:ux-design
 argument-hint: "[product-name]"
-allowed-tools: ["Read", "Edit", "Write", "Glob", "Grep", "Task"]
+allowed-tools: ["Read", "Edit", "Write", "Glob", "Grep", "Task", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:design-system skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:design-system")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.

commands/work-on-next-task.md

@@ -1,8 +1,11 @@
 ---
 name: groundwork:work-on-next-task
 description: Execute next uncompleted task with full project context (PRD, architecture, tasks). Usage /groundwork:work-on-next-task
-allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Task"]
+allowed-tools: ["Read", "Edit", "Write", "Bash", "Glob", "Grep", "Task", "Skill"]
 disable-model-invocation: true
 ---
 
-Invoke the groundwork:next-task skill and follow it exactly as presented to you
+CRITICAL INSTRUCTION: Before doing ANYTHING else, you MUST call the Skill tool with:
+  Skill(skill="groundwork:next-task")
+
+Do NOT read files, explore code, or generate any response before invoking this skill. The skill contains your complete workflow and you must follow it exactly as presented to you.