Fix Deacon spin indefinitely bug (hq-oosxt)

- Add heartbeat checking to Boot degraded triage: detects stale Deacon
  heartbeat (>15min nudges, >30min restarts)
- Add checkDeaconHeartbeat to daemon heartbeat cycle as fallback
- This ensures the Deacon is monitored continuously, not just at startup

The mol-deacon-patrol formula was also updated separately to use
gt deacon health-check instead of ephemeral context memory tracking.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

internal/cmd/boot.go

@@ -8,6 +8,7 @@ import (
 
 	"github.com/spf13/cobra"
 	"github.com/steveyegge/gastown/internal/boot"
+	"github.com/steveyegge/gastown/internal/deacon"
 	"github.com/steveyegge/gastown/internal/style"
 	"github.com/steveyegge/gastown/internal/workspace"
 )
@@ -288,9 +289,30 @@ func runDegradedTriage(b *boot.Boot) (action, target string, err error) {
 		return "report", "deacon-missing", nil
 	}
 
-	// Deacon exists - check if it's responsive (basic pane output check)
-	// In degraded mode, we can't do sophisticated analysis
-	// Just verify the session is alive
+	// Deacon exists - check heartbeat to detect stuck sessions
+	// A session can exist but be stuck (not making progress)
+	townRoot, _ := workspace.FindFromCwd()
+	if townRoot != "" {
+		hb := deacon.ReadHeartbeat(townRoot)
+		if hb.ShouldPoke() {
+			// Heartbeat is stale (>15 min) - Deacon is stuck
+			// Nudge the session to try to wake it up
+			age := hb.Age()
+			if age > 30*time.Minute {
+				// Very stuck - restart the session
+				fmt.Printf("Deacon heartbeat is %s old - restarting session\n", age.Round(time.Minute))
+				if err := tm.KillSession(deaconSession); err == nil {
+					return "restart", "deacon-stuck", nil
+				}
+			} else {
+				// Stuck but not critically - try nudging first
+				fmt.Printf("Deacon heartbeat is %s old - nudging session\n", age.Round(time.Minute))
+				_ = tm.NudgeSession(deaconSession, "HEALTH_CHECK: heartbeat is stale, respond to confirm responsiveness")
+				return "nudge", "deacon-stale", nil
+			}
+		}
+	}
+
 	return "nothing", "", nil
 }
 

internal/daemon/daemon.go

@@ -17,6 +17,7 @@ import (
 	"github.com/steveyegge/gastown/internal/boot"
 	"github.com/steveyegge/gastown/internal/config"
 	"github.com/steveyegge/gastown/internal/constants"
+	"github.com/steveyegge/gastown/internal/deacon"
 	"github.com/steveyegge/gastown/internal/feed"
 	"github.com/steveyegge/gastown/internal/polecat"
 	"github.com/steveyegge/gastown/internal/tmux"
@@ -148,6 +149,10 @@ func (d *Daemon) heartbeat(state *State) {
 	// Boot handles the "when to wake Deacon" decision via triage logic
 	d.ensureBootRunning()
 
+	// 1b. Direct Deacon heartbeat check (belt-and-suspenders)
+	// Boot may not detect all stuck states; this provides a fallback
+	d.checkDeaconHeartbeat()
+
 	// 2. Ensure Witnesses are running for all rigs (restart if dead)
 	d.ensureWitnessesRunning()
 
@@ -293,6 +298,54 @@ func (d *Daemon) ensureDeaconRunning() {
 	d.logger.Println("Deacon session started successfully")
 }
 
+// checkDeaconHeartbeat checks if the Deacon is making progress.
+// This is a belt-and-suspenders fallback in case Boot doesn't detect stuck states.
+// Uses the heartbeat file that the Deacon updates on each patrol cycle.
+func (d *Daemon) checkDeaconHeartbeat() {
+	hb := deacon.ReadHeartbeat(d.config.TownRoot)
+	if hb == nil {
+		// No heartbeat file - Deacon hasn't started a cycle yet
+		return
+	}
+
+	age := hb.Age()
+
+	// If heartbeat is very stale (>15 min), the Deacon is likely stuck
+	if !hb.ShouldPoke() {
+		// Heartbeat is fresh enough
+		return
+	}
+
+	d.logger.Printf("Deacon heartbeat is stale (%s old), checking session...", age.Round(time.Minute))
+
+	// Check if session exists
+	hasSession, err := d.tmux.HasSession(DeaconSessionName)
+	if err != nil {
+		d.logger.Printf("Error checking Deacon session: %v", err)
+		return
+	}
+
+	if !hasSession {
+		// Session doesn't exist - ensureBootRunning will handle restart
+		return
+	}
+
+	// Session exists but heartbeat is stale - Deacon is stuck
+	if age > 30*time.Minute {
+		// Very stuck - restart the session
+		d.logger.Printf("Deacon stuck for %s - restarting session", age.Round(time.Minute))
+		if err := d.tmux.KillSession(DeaconSessionName); err != nil {
+			d.logger.Printf("Error killing stuck Deacon: %v", err)
+		}
+		// ensureDeaconRunning will be called next heartbeat to restart
+	} else {
+		// Stuck but not critically - nudge to wake up
+		d.logger.Printf("Deacon stuck for %s - nudging session", age.Round(time.Minute))
+		if err := d.tmux.NudgeSession(DeaconSessionName, "HEALTH_CHECK: heartbeat stale, respond to confirm responsiveness"); err != nil {
+			d.logger.Printf("Error nudging stuck Deacon: %v", err)
+		}
+	}
+}
 
 // ensureWitnessesRunning ensures witnesses are running for all rigs.
 // Called on each heartbeat to maintain witness patrol loops.