propagation: single registry drives all node references

New bin/netvm-registry.py parses NODES.md (node -> cdp_port); every
consumer reads from it instead of hardcoding:
- onboard-driver.py: CDP_PORTS dict -> registry lookup (new nodes work)
- muse-signin.py: hardcoded 9410 -> --node/--cdp-port args
- muse-chat-api.py: hardcoded ACCOUNTS -> registry-built
- agent-health.sh: hardcoded muse/pip blocks -> loop over all active
  nodes (646 and opm now get health coverage too)
NODES.md: record opm node (9440).

bin/agent-health.sh

@@ -49,28 +49,27 @@ restart_browser() {
 # Main
 echo "=== Health check $(date -Iseconds) ===" >> "$LOG"
 
-# muse (node muse, CDP 9410)
-if ! check_agent "muse" "muse" "9410"; then
-    restart_browser "muse" "9410"
-    sleep 5
-    if ! check_agent "muse" "muse" "9410"; then
-        echo "$(date -Iseconds) muse: CRITICAL - still down after restart" >> "$LOG"
-        # TODO: Alert operator (e.g., via board post or email)
-    else
-        echo "$(date -Iseconds) muse: RECOVERED after restart" >> "$LOG"
+check_one() {
+    local agent=$1
+    local cdp_port=$2
+    # node == agent == profile (unified naming)
+    if ! check_agent "$agent" "$agent" "$cdp_port"; then
+        restart_browser "$agent" "$cdp_port"
+        sleep 5
+        if ! check_agent "$agent" "$agent" "$cdp_port"; then
+            echo "$(date -Iseconds) $agent: CRITICAL - still down after restart" >> "$LOG"
+            # TODO: Alert operator (e.g., via board post or email)
+        else
+            echo "$(date -Iseconds) $agent: RECOVERED after restart" >> "$LOG"
+        fi
     fi
-fi
+}
 
-# pip (node pip, CDP 9420)
-if ! check_agent "pip" "pip" "9420"; then
-    restart_browser "pip" "9420"
-    sleep 5
-    if ! check_agent "pip" "pip" "9420"; then
-        echo "$(date -Iseconds) pip: CRITICAL - still down after restart" >> "$LOG"
-    else
-        echo "$(date -Iseconds) pip: RECOVERED after restart" >> "$LOG"
-    fi
-fi
+# Every active node in the NODES.md registry gets checked — new nodes
+# propagate automatically, no per-node blocks to add.
+"$NETVM_BIN/netvm-registry.py" 2>/dev/null | while IFS=: read -r node port; do
+    [ -n "$node" ] && [ -n "$port" ] && check_one "$node" "$port"
+done
 
 # Trim log (keep last 1000 lines)
 tail -1000 "$LOG" > "$LOG.tmp" && mv "$LOG.tmp" "$LOG"