Replace invalid hide-from-slash-command-tool with official frontmatter attrs

- Cancel commands: use disable-model-invocation to prevent agent self-cancel
- Duplicate skills: add disable-model-invocation to hide from Claude Code
- Remove non-functional hide-from-slash-command-tool from start/gen-plan commands
- Rename strip function in install-skill.sh to strip all Claude-specific attrs

Author: SihaoLiu

commands/cancel-pr-loop.md

@@ -1,7 +1,7 @@
 ---
 description: "Cancel active PR loop"
 allowed-tools: ["Bash(${CLAUDE_PLUGIN_ROOT}/scripts/cancel-pr-loop.sh)", "Bash(${CLAUDE_PLUGIN_ROOT}/scripts/cancel-pr-loop.sh --force)"]
-hide-from-slash-command-tool: "true"
+disable-model-invocation: true
 ---
 
 # Cancel PR Loop

commands/cancel-rlcr-loop.md

@@ -1,7 +1,7 @@
 ---
 description: "Cancel active RLCR loop"
 allowed-tools: ["Bash(${CLAUDE_PLUGIN_ROOT}/scripts/cancel-rlcr-loop.sh)", "Bash(${CLAUDE_PLUGIN_ROOT}/scripts/cancel-rlcr-loop.sh --force)", "AskUserQuestion"]
-hide-from-slash-command-tool: "true"
+disable-model-invocation: true
 ---
 
 # Cancel RLCR Loop

commands/gen-plan.md

@@ -11,7 +11,6 @@ allowed-tools:
   - "Task"
   - "Write"
   - "AskUserQuestion"
-hide-from-slash-command-tool: "true"
 ---
 
 # Generate Plan from Draft

commands/start-pr-loop.md

@@ -2,7 +2,6 @@
 description: "Start PR review loop with bot monitoring"
 argument-hint: "--claude|--codex [--max N] [--codex-model MODEL:EFFORT] [--codex-timeout SECONDS]"
 allowed-tools: ["Bash(${CLAUDE_PLUGIN_ROOT}/scripts/setup-pr-loop.sh:*)"]
-hide-from-slash-command-tool: "true"
 ---
 
 # Start PR Loop

commands/start-rlcr-loop.md

@@ -6,7 +6,6 @@ allowed-tools:
   - "Read"
   - "Task"
   - "AskUserQuestion"
-hide-from-slash-command-tool: "true"
 ---
 
 # Start RLCR Loop

scripts/install-skill.sh

@@ -151,7 +151,7 @@ hydrate_skill_runtime_root() {
     done
 }
 
-strip_user_invocable_for_runtime() {
+strip_claude_specific_frontmatter() {
     local target_dir="$1"
     local skill
     local skill_file
@@ -162,7 +162,7 @@ strip_user_invocable_for_runtime() {
         [[ -f "$skill_file" ]] || continue
 
         if [[ "$DRY_RUN" == "true" ]]; then
-            log "DRY-RUN strip user-invocable in $skill_file"
+            log "DRY-RUN strip Claude-specific frontmatter in $skill_file"
             continue
         fi
 
@@ -180,6 +180,8 @@ strip_user_invocable_for_runtime() {
                 next
             }
             in_fm && $0 ~ /^user-invocable:[[:space:]]*/ { next }
+            in_fm && $0 ~ /^disable-model-invocation:[[:space:]]*/ { next }
+            in_fm && $0 ~ /^hide-from-slash-command-tool:[[:space:]]*/ { next }
             { print }
         ' "$skill_file" > "$tmp" \
             || { rm -f "$tmp"; die "failed to update $skill_file"; }
@@ -204,7 +206,7 @@ sync_target() {
     done
     install_runtime_bundle "$target_dir"
     hydrate_skill_runtime_root "$target_dir"
-    strip_user_invocable_for_runtime "$target_dir"
+    strip_claude_specific_frontmatter "$target_dir"
 }
 
 while [[ $# -gt 0 ]]; do

skills/humanize-gen-plan/SKILL.md

@@ -3,6 +3,7 @@ name: humanize-gen-plan
 description: Generate a structured implementation plan from a draft document. Validates input, checks relevance, analyzes for issues, and generates a complete plan.md with acceptance criteria.
 type: flow
 user-invocable: false
+disable-model-invocation: true
 ---
 
 # Humanize Generate Plan

skills/humanize-rlcr/SKILL.md

@@ -3,6 +3,7 @@ name: humanize-rlcr
 description: Start RLCR (Ralph-Loop with Codex Review) with hook-equivalent enforcement from skill mode by reusing the existing stop-hook logic.
 type: flow
 user-invocable: false
+disable-model-invocation: true
 ---
 
 # Humanize RLCR Loop (Hook-Equivalent)

skills/humanize/SKILL.md

@@ -2,6 +2,7 @@
 name: humanize
 description: Iterative development with AI review. Provides RLCR (Ralph-Loop with Codex Review) for implementation planning and code review loops, plus PR review automation with bot monitoring.
 user-invocable: false
+disable-model-invocation: true
 ---
 
 # Humanize - Iterative Development with AI Review