style(no-slop): remove every em-dash + banned words across all modules + capstone

Apply the no-ai-slop standard (now binding in AGENTS.md): the em-dash character is
banned outright (restructured, not blind-replaced), plus the banned word/phrase
list (delve, leverage, robust, seamless, truly, unlock, etc.). 0 em-dashes remain
in modules + capstone; the only "robust" left is the planted M10 ai-change.patch
trap. Module H1 titles use a colon separator.

All deliberate teaching devices preserved; labs compile/parse (py/sh/yaml/json);
no junk. AGENTS.md updated with the hard no-slop rules.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TfzV5QvtPDz8LJS3Pu5VLT

modules/22-securing-third-party-mcp-and-skills/lab/audit.sh

@@ -1,10 +1,10 @@
 #!/usr/bin/env bash
 #
-# audit.sh — a runnable version of the Module 22 vetting checklist.
+# audit.sh: a runnable version of the Module 22 vetting checklist.
 #
 # Static red-flag scan over a third-party MCP server or skill BEFORE you install it. It does not
 # execute anything in the target; it only reads. A clean run is NOT a guarantee (see "Where it
-# breaks") — it is a cheap first pass that catches the obvious and the lazy.
+# breaks"); it is a cheap first pass that catches the obvious and the lazy.
 #
 # Usage:  bash audit.sh <path-to-skill-or-server-dir>
 #
@@ -19,7 +19,7 @@ fi
 hits=0
 section () { printf '\n=== %s ===\n' "$1"; }
 
-# scan <label> <regex>  — grep the tree, print matches, count a hit if found
+# scan <label> <regex>: grep the tree, print matches, count a hit if found
 scan () {
   local label="$1" regex="$2" out
   out=$(grep -rIinE "$regex" "$TARGET" 2>/dev/null || true)
@@ -79,7 +79,7 @@ fi
 
 section "Verdict"
 if (( hits > 0 )); then
-  echo "REJECT (or sandbox + scope) — $hits red-flag categor$([[ $hits -eq 1 ]] && echo y || echo ies) tripped."
+  echo "REJECT (or sandbox + scope): $hits red-flag categor$([[ $hits -eq 1 ]] && echo y || echo ies) tripped."
   echo "Read the flagged lines above against what the skill CLAIMS to do."
   exit 1
 else