diff --git a/pkg/agent/agent_command.go b/pkg/agent/agent_command.go
index 944c58f3d..efd5326cf 100644
--- a/pkg/agent/agent_command.go
+++ b/pkg/agent/agent_command.go
@@ -350,11 +350,12 @@ func (al *AgentLoop) buildCommandsRuntime(
 			}
 			history := agent.Sessions.GetHistory(opts.SessionKey)
 			return &commands.ContextStats{
-				UsedTokens:       usage.UsedTokens,
-				TotalTokens:      usage.TotalTokens,
-				CompressAtTokens: usage.CompressAtTokens,
-				UsedPercent:      usage.UsedPercent,
-				MessageCount:     len(history),
+				UsedTokens:        usage.UsedTokens,
+				TotalTokens:       usage.TotalTokens,
+				CompressAtTokens:  usage.CompressAtTokens,
+				SummarizeAtTokens: usage.SummarizeAtTokens,
+				UsedPercent:       usage.UsedPercent,
+				MessageCount:      len(history),
 			}
 		}
 	}
diff --git a/pkg/agent/context_usage.go b/pkg/agent/context_usage.go
index 39d4f3dee..9b9222401 100644
--- a/pkg/agent/context_usage.go
+++ b/pkg/agent/context_usage.go
@@ -61,6 +61,13 @@ func computeContextUsage(agent *AgentInstance, sessionKey string) *bus.ContextUs
 	// proactive trigger (msgTokens + toolTokens + maxTokens > contextWindow).
 	compressAt := effectiveWindow
 
+	// summarizeAt = soft summarization trigger: matches maybeSummarize's
+	// threshold (contextWindow * SummarizeTokenPercent / 100).
+	summarizeAt := contextWindow * agent.SummarizeTokenPercent / 100
+	if summarizeAt <= 0 {
+		summarizeAt = compressAt
+	}
+
 	usedPercent := 0
 	if compressAt > 0 {
 		usedPercent = usedTokens * 100 / compressAt
@@ -70,9 +77,10 @@ func computeContextUsage(agent *AgentInstance, sessionKey string) *bus.ContextUs
 	}
 
 	return &bus.ContextUsage{
-		UsedTokens:       usedTokens,
-		TotalTokens:      contextWindow,
-		CompressAtTokens: compressAt,
-		UsedPercent:      usedPercent,
+		UsedTokens:        usedTokens,
+		TotalTokens:       contextWindow,
+		CompressAtTokens:  compressAt,
+		SummarizeAtTokens: summarizeAt,
+		UsedPercent:       usedPercent,
 	}
 }
diff --git a/pkg/bus/types.go b/pkg/bus/types.go
index 953e69d9c..6863a49ec 100644
--- a/pkg/bus/types.go
+++ b/pkg/bus/types.go
@@ -64,10 +64,11 @@ type OutboundScope struct {
 // ContextUsage describes how much of the model's context window the current
 // session consumes, and how far it is from triggering compression.
 type ContextUsage struct {
-	UsedTokens       int `json:"used_tokens"`
-	TotalTokens      int `json:"total_tokens"`       // model context window
-	CompressAtTokens int `json:"compress_at_tokens"` // threshold that triggers compression
-	UsedPercent      int `json:"used_percent"`       // 0-100
+	UsedTokens        int `json:"used_tokens"`
+	TotalTokens       int `json:"total_tokens"`        // model context window
+	CompressAtTokens  int `json:"compress_at_tokens"`  // hard budget compression threshold (contextWindow - maxTokens)
+	SummarizeAtTokens int `json:"summarize_at_tokens"` // soft summarization trigger (contextWindow * summarizeTokenPercent / 100)
+	UsedPercent       int `json:"used_percent"`        // 0-100, relative to compressAt
 }
 
 type OutboundMessage struct {
diff --git a/pkg/channels/pico/pico.go b/pkg/channels/pico/pico.go
index 3c3938989..adceecad7 100644
--- a/pkg/channels/pico/pico.go
+++ b/pkg/channels/pico/pico.go
@@ -1394,10 +1394,11 @@ func setContextUsagePayload(payload map[string]any, u *bus.ContextUsage) {
 		return
 	}
 	payload["context_usage"] = map[string]any{
-		"used_tokens":        u.UsedTokens,
-		"total_tokens":       u.TotalTokens,
-		"compress_at_tokens": u.CompressAtTokens,
-		"used_percent":       u.UsedPercent,
+		"used_tokens":         u.UsedTokens,
+		"total_tokens":        u.TotalTokens,
+		"compress_at_tokens":  u.CompressAtTokens,
+		"summarize_at_tokens": u.SummarizeAtTokens,
+		"used_percent":        u.UsedPercent,
 	}
 }
 
diff --git a/pkg/commands/cmd_context.go b/pkg/commands/cmd_context.go
index 55481662c..4b4eddfa6 100644
--- a/pkg/commands/cmd_context.go
+++ b/pkg/commands/cmd_context.go
@@ -29,14 +29,16 @@ func formatContextStats(s *ContextStats) string {
 		remaining = 0
 	}
 	usedWindowPercent := s.UsedTokens * 100 / max(s.TotalTokens, 1)
-	return fmt.Sprintf(
-		"Context usage  \nMessages: %d  \nUsed: ~%d / %d tokens (%d%%)  \nCompress at: %d tokens  \nCompression progress: %d%%  \nRemaining: ~%d tokens",
+	msg := fmt.Sprintf(
+		"Context usage  \nMessages: %d  \nUsed: ~%d / %d tokens (%d%%)  \nCompress at: %d tokens  \nSummarize at: %d tokens  \nCompression progress: %d%%  \nRemaining: ~%d tokens",
 		s.MessageCount,
 		s.UsedTokens,
 		s.TotalTokens,
 		usedWindowPercent,
 		s.CompressAtTokens,
+		s.SummarizeAtTokens,
 		s.UsedPercent,
 		remaining,
 	)
+	return msg
 }
diff --git a/pkg/commands/runtime.go b/pkg/commands/runtime.go
index b0327c863..ba7d003c4 100644
--- a/pkg/commands/runtime.go
+++ b/pkg/commands/runtime.go
@@ -29,11 +29,12 @@ type MCPToolInfo struct {
 
 // ContextStats describes current session context window usage.
 type ContextStats struct {
-	UsedTokens       int
-	TotalTokens      int // model context window
-	CompressAtTokens int // compression threshold
-	UsedPercent      int // 0-100
-	MessageCount     int
+	UsedTokens        int
+	TotalTokens       int // model context window
+	CompressAtTokens  int // hard budget compression threshold
+	SummarizeAtTokens int // soft summarization trigger
+	UsedPercent       int // 0-100
+	MessageCount      int
 }
 
 // StopResult describes the outcome of a stop request for the current session.
diff --git a/web/frontend/src/components/chat/context-usage-ring.tsx b/web/frontend/src/components/chat/context-usage-ring.tsx
index 037a20cef..7a33dc2fe 100644
--- a/web/frontend/src/components/chat/context-usage-ring.tsx
+++ b/web/frontend/src/components/chat/context-usage-ring.tsx
@@ -145,6 +145,23 @@ export function ContextUsageRing({
             />
           </div>
 
+          <div className="mt-2 space-y-0.5">
+            <div className="flex items-center justify-between text-[10px]">
+              <span className="text-muted-foreground">Compress at</span>
+              <span className="tabular-nums">
+                {formatTokens(usage.compress_at_tokens)}
+              </span>
+            </div>
+            {usage.summarize_at_tokens != null && usage.summarize_at_tokens > 0 && (
+              <div className="flex items-center justify-between text-[10px]">
+                <span className="text-muted-foreground">Summarize at</span>
+                <span className="tabular-nums">
+                  {formatTokens(usage.summarize_at_tokens)}
+                </span>
+              </div>
+            )}
+          </div>
+
           <button
             type="button"
             onClick={handleDetail}
diff --git a/web/frontend/src/features/chat/protocol.ts b/web/frontend/src/features/chat/protocol.ts
index b372b235c..59eff92cd 100644
--- a/web/frontend/src/features/chat/protocol.ts
+++ b/web/frontend/src/features/chat/protocol.ts
@@ -79,6 +79,7 @@ function parseContextUsage(
     used_tokens: used,
     total_tokens: total,
     compress_at_tokens: Number(obj.compress_at_tokens) || 0,
+    summarize_at_tokens: obj.summarize_at_tokens != null ? Number(obj.summarize_at_tokens) : undefined,
     used_percent: Number(obj.used_percent) || 0,
   }
 }
diff --git a/web/frontend/src/store/chat.ts b/web/frontend/src/store/chat.ts
index 056d5601b..7c605b13f 100644
--- a/web/frontend/src/store/chat.ts
+++ b/web/frontend/src/store/chat.ts
@@ -53,6 +53,7 @@ export interface ContextUsage {
   used_tokens: number
   total_tokens: number
   compress_at_tokens: number
+  summarize_at_tokens?: number
   used_percent: number
 }