diff --git a/.goreleaser.yaml b/.goreleaser.yaml
index 0f2add2..5f28d54 100644
--- a/.goreleaser.yaml
+++ b/.goreleaser.yaml
@@ -12,6 +12,13 @@ before:
     - bash -c "cd ui && npm ci --silent && npm run build"
     - bash -c "rm -rf internal/worker/static && mkdir -p internal/worker/static && cp -r ui/dist/* internal/worker/static/"
 
+# Generate versioned plugin configuration files (GoReleaser Pro feature)
+template_files:
+  - src: plugin/.claude-plugin/plugin.json.tpl
+    dst: .claude-plugin/plugin.json
+  - src: plugin/.claude-plugin/marketplace.json.tpl
+    dst: .claude-plugin/marketplace.json
+
 builds:
   # Worker service
   - id: worker
@@ -211,7 +218,7 @@ archives:
       {{- .Os }}_
       {{- .Arch }}
     files:
-      - src: plugin/.claude-plugin/*
+      - src: dist/.claude-plugin/*
         dst: .claude-plugin
         strip_parent: true
       - src: plugin/hooks/hooks.json
diff --git a/Makefile b/Makefile
index 9f33b5f..f8a7cb0 100644
--- a/Makefile
+++ b/Makefile
@@ -45,6 +45,7 @@ hooks:
 	go build $(BUILD_TAGS) $(LDFLAGS) -o $(BUILD_DIR)/hooks/post-tool-use ./cmd/hooks/post-tool-use
 	go build $(BUILD_TAGS) $(LDFLAGS) -o $(BUILD_DIR)/hooks/subagent-stop ./cmd/hooks/subagent-stop
 	go build $(BUILD_TAGS) $(LDFLAGS) -o $(BUILD_DIR)/hooks/stop ./cmd/hooks/stop
+	go build $(BUILD_TAGS) $(LDFLAGS) -o $(BUILD_DIR)/hooks/statusline ./cmd/hooks/statusline
 
 # Build MCP server
 mcp:
@@ -65,6 +66,7 @@ build-linux:
 	GOOS=linux GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/linux-amd64/hooks/post-tool-use ./cmd/hooks/post-tool-use
 	GOOS=linux GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/linux-amd64/hooks/subagent-stop ./cmd/hooks/subagent-stop
 	GOOS=linux GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/linux-amd64/hooks/stop ./cmd/hooks/stop
+	GOOS=linux GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/linux-amd64/hooks/statusline ./cmd/hooks/statusline
 
 build-darwin:
 	@echo "Building for macOS..."
@@ -83,6 +85,8 @@ build-darwin:
 	GOOS=darwin GOARCH=arm64 go build $(LDFLAGS) -o $(BUILD_DIR)/darwin-arm64/hooks/subagent-stop ./cmd/hooks/subagent-stop
 	GOOS=darwin GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/darwin-amd64/hooks/stop ./cmd/hooks/stop
 	GOOS=darwin GOARCH=arm64 go build $(LDFLAGS) -o $(BUILD_DIR)/darwin-arm64/hooks/stop ./cmd/hooks/stop
+	GOOS=darwin GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/darwin-amd64/hooks/statusline ./cmd/hooks/statusline
+	GOOS=darwin GOARCH=arm64 go build $(LDFLAGS) -o $(BUILD_DIR)/darwin-arm64/hooks/statusline ./cmd/hooks/statusline
 
 build-windows:
 	@echo "Building for Windows..."
@@ -94,6 +98,7 @@ build-windows:
 	GOOS=windows GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/windows-amd64/hooks/post-tool-use.exe ./cmd/hooks/post-tool-use
 	GOOS=windows GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/windows-amd64/hooks/subagent-stop.exe ./cmd/hooks/subagent-stop
 	GOOS=windows GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/windows-amd64/hooks/stop.exe ./cmd/hooks/stop
+	GOOS=windows GOARCH=amd64 go build $(LDFLAGS) -o $(BUILD_DIR)/windows-amd64/hooks/statusline.exe ./cmd/hooks/statusline
 
 # Stop any running worker
 stop-worker:
@@ -107,8 +112,8 @@ stop-worker:
 start-worker:
 	@echo "Starting worker..."
 	@# Prefer cache directory (where Claude Code looks), fall back to marketplaces
-	@if [ -f "$(HOME)/.claude/plugins/cache/claude-mnemonic/claude-mnemonic/1.0.0/worker" ]; then \
-		nohup $(HOME)/.claude/plugins/cache/claude-mnemonic/claude-mnemonic/1.0.0/worker > /tmp/claude-mnemonic-worker.log 2>&1 & \
+	@if [ -f "$(HOME)/.claude/plugins/cache/claude-mnemonic/claude-mnemonic/$(VERSION)/worker" ]; then \
+		nohup $(HOME)/.claude/plugins/cache/claude-mnemonic/claude-mnemonic/$(VERSION)/worker > /tmp/claude-mnemonic-worker.log 2>&1 & \
 	else \
 		nohup $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/worker > /tmp/claude-mnemonic-worker.log 2>&1 & \
 	fi
@@ -132,22 +137,11 @@ install: build stop-worker
 	cp $(BUILD_DIR)/mcp-server $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/
 	cp $(BUILD_DIR)/hooks/* $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/hooks/
 	cp $(PLUGIN_DIR)/hooks/hooks.json $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/hooks/
-	cp $(PLUGIN_DIR)/.claude-plugin/plugin.json $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/.claude-plugin/
-	cp $(PLUGIN_DIR)/.claude-plugin/marketplace.json $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/.claude-plugin/
-	@# Also install to cache directory (where Claude Code looks for plugins)
-	@if [ -d "$(HOME)/.claude/plugins/cache/claude-mnemonic" ]; then \
-		echo "Updating plugin in cache directory..."; \
-		CACHE_DIR=$$(find $(HOME)/.claude/plugins/cache/claude-mnemonic -type d -name "hooks" -exec dirname {} \; 2>/dev/null | head -1); \
-		if [ -n "$$CACHE_DIR" ]; then \
-			cp $(BUILD_DIR)/worker "$$CACHE_DIR/"; \
-			cp $(BUILD_DIR)/mcp-server "$$CACHE_DIR/"; \
-			cp $(BUILD_DIR)/hooks/* "$$CACHE_DIR/hooks/"; \
-			cp $(PLUGIN_DIR)/hooks/hooks.json "$$CACHE_DIR/hooks/"; \
-			echo "Cache directory updated: $$CACHE_DIR"; \
-		fi; \
-	fi
+	@# Update plugin.json and marketplace.json with current version to prevent stale version directories
+	@sed 's/"version": "[^"]*"/"version": "$(VERSION)"/g' $(PLUGIN_DIR)/.claude-plugin/plugin.json > $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/.claude-plugin/plugin.json
+	@sed 's/"version": "[^"]*"/"version": "$(VERSION)"/g' $(PLUGIN_DIR)/.claude-plugin/marketplace.json > $(HOME)/.claude/plugins/marketplaces/claude-mnemonic/.claude-plugin/marketplace.json
 	@echo "Registering plugin with Claude Code..."
-	@./scripts/register-plugin.sh
+	@./scripts/register-plugin.sh "$(VERSION)"
 	@$(MAKE) start-worker
 	@echo "Installation complete!"
 
diff --git a/README.md b/README.md
index cd90498..38af3e9 100644
--- a/README.md
+++ b/README.md
@@ -100,6 +100,7 @@ cosign verify-blob \
 | **Project Isolation** | Each project has its own knowledge base |
 | **Global Patterns** | Best practices are shared across all projects |
 | **Semantic Search** | Find relevant context with natural language |
+| **Live Statusline** | Real-time metrics in Claude Code: `[mnemonic] ● served:42 | project:28 memories` |
 | **Web Dashboard** | Browse and manage memories at `localhost:37777` |
 
 ### How knowledge flows
diff --git a/cmd/hooks/statusline/main.go b/cmd/hooks/statusline/main.go
new file mode 100644
index 0000000..6876d84
--- /dev/null
+++ b/cmd/hooks/statusline/main.go
@@ -0,0 +1,307 @@
+// Package main provides the statusline hook for Claude Code.
+// This binary outputs a status line showing claude-mnemonic metrics.
+package main
+
+import (
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"os"
+	"time"
+
+	"github.com/lukaszraczylo/claude-mnemonic/pkg/hooks"
+)
+
+// StatusInput is the JSON input from Claude Code's statusline feature.
+type StatusInput struct {
+	HookEventName string `json:"hook_event_name"`
+	SessionID     string `json:"session_id"`
+	CWD           string `json:"cwd"`
+	Model         struct {
+		ID          string `json:"id"`
+		DisplayName string `json:"display_name"`
+	} `json:"model"`
+	Workspace struct {
+		CurrentDir string `json:"current_dir"`
+		ProjectDir string `json:"project_dir"`
+	} `json:"workspace"`
+	Version string `json:"version"`
+	Cost    struct {
+		TotalCostUSD       float64 `json:"total_cost_usd"`
+		TotalDurationMS    int64   `json:"total_duration_ms"`
+		TotalAPIDurationMS int64   `json:"total_api_duration_ms"`
+		TotalLinesAdded    int     `json:"total_lines_added"`
+		TotalLinesRemoved  int     `json:"total_lines_removed"`
+	} `json:"cost"`
+	ContextWindow struct {
+		TotalInputTokens  int `json:"total_input_tokens"`
+		TotalOutputTokens int `json:"total_output_tokens"`
+		ContextWindowSize int `json:"context_window_size"`
+	} `json:"context_window"`
+}
+
+// WorkerStats is the response from the worker's /api/stats endpoint.
+type WorkerStats struct {
+	Uptime              string `json:"uptime"`
+	ActiveSessions      int    `json:"activeSessions"`
+	QueueDepth          int    `json:"queueDepth"`
+	IsProcessing        bool   `json:"isProcessing"`
+	ConnectedClients    int    `json:"connectedClients"`
+	SessionsToday       int    `json:"sessionsToday"`
+	Ready               bool   `json:"ready"`
+	Project             string `json:"project,omitempty"`
+	ProjectObservations int    `json:"projectObservations,omitempty"`
+	Retrieval           struct {
+		TotalRequests      int64 `json:"TotalRequests"`
+		ObservationsServed int64 `json:"ObservationsServed"`
+		SearchRequests     int64 `json:"SearchRequests"`
+		ContextInjections  int64 `json:"ContextInjections"`
+	} `json:"retrieval"`
+}
+
+// ANSI color codes
+const (
+	colorReset  = "\033[0m"
+	colorGreen  = "\033[32m"
+	colorYellow = "\033[33m"
+	colorCyan   = "\033[36m"
+	colorGray   = "\033[90m"
+	colorRed    = "\033[31m"
+)
+
+func main() {
+	// Read input from stdin
+	inputData, err := io.ReadAll(os.Stdin)
+	if err != nil {
+		// On error, output minimal status
+		fmt.Println(formatOffline())
+		return
+	}
+
+	var input StatusInput
+	if err := json.Unmarshal(inputData, &input); err != nil {
+		fmt.Println(formatOffline())
+		return
+	}
+
+	// Determine project directory
+	projectDir := input.Workspace.ProjectDir
+	if projectDir == "" {
+		projectDir = input.Workspace.CurrentDir
+	}
+	if projectDir == "" {
+		projectDir = input.CWD
+	}
+
+	// Generate project ID
+	project := ""
+	if projectDir != "" {
+		project = hooks.ProjectIDWithName(projectDir)
+	}
+
+	// Get worker stats
+	stats := getWorkerStats(project)
+
+	// Format and output statusline
+	fmt.Println(formatStatusLine(stats, input))
+}
+
+// getWorkerStats fetches stats from the worker service.
+func getWorkerStats(project string) *WorkerStats {
+	port := hooks.GetWorkerPort()
+
+	// Build URL with optional project parameter
+	endpoint := fmt.Sprintf("http://127.0.0.1:%d/api/stats", port)
+	if project != "" {
+		endpoint += "?project=" + url.QueryEscape(project)
+	}
+
+	// Create HTTP client with short timeout (statusline must be fast)
+	client := &http.Client{Timeout: 100 * time.Millisecond}
+
+	resp, err := client.Get(endpoint)
+	if err != nil {
+		return nil
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		return nil
+	}
+
+	var stats WorkerStats
+	if err := json.NewDecoder(resp.Body).Decode(&stats); err != nil {
+		return nil
+	}
+
+	return &stats
+}
+
+// formatStatusLine formats the status line output.
+func formatStatusLine(stats *WorkerStats, input StatusInput) string {
+	// Check if colors are enabled (default: yes, unless TERM is dumb or NO_COLOR is set)
+	useColors := os.Getenv("NO_COLOR") == "" && os.Getenv("TERM") != "dumb"
+	if os.Getenv("CLAUDE_MNEMONIC_STATUSLINE_COLORS") == "false" {
+		useColors = false
+	} else if os.Getenv("CLAUDE_MNEMONIC_STATUSLINE_COLORS") == "true" {
+		useColors = true
+	}
+
+	// Check format preference
+	format := os.Getenv("CLAUDE_MNEMONIC_STATUSLINE_FORMAT")
+	if format == "" {
+		format = "default"
+	}
+
+	if stats == nil {
+		return formatOfflineColored(useColors)
+	}
+
+	if !stats.Ready {
+		return formatStartingColored(useColors)
+	}
+
+	switch format {
+	case "compact":
+		return formatCompact(stats, useColors)
+	case "minimal":
+		return formatMinimal(stats, useColors)
+	default:
+		return formatDefault(stats, useColors)
+	}
+}
+
+// formatDefault returns the default status line format.
+func formatDefault(stats *WorkerStats, useColors bool) string {
+	// [mnemonic] ● served:42 | injected:5 | searches:3 | project:28 memories
+	var prefix, indicator, reset string
+	if useColors {
+		prefix = colorCyan + "[mnemonic]" + colorReset
+		indicator = colorGreen + "●" + colorReset
+		reset = colorReset
+	} else {
+		prefix = "[mnemonic]"
+		indicator = "●"
+	}
+
+	// Build status parts with clear labels
+	parts := []string{}
+
+	// Total memories served to Claude this session
+	parts = append(parts, fmt.Sprintf("served:%d", stats.Retrieval.ObservationsServed))
+
+	// Context injections (memories auto-loaded at session start)
+	if stats.Retrieval.ContextInjections > 0 {
+		parts = append(parts, fmt.Sprintf("injected:%d", stats.Retrieval.ContextInjections))
+	}
+
+	// Semantic searches performed
+	if stats.Retrieval.SearchRequests > 0 {
+		parts = append(parts, fmt.Sprintf("searches:%d", stats.Retrieval.SearchRequests))
+	}
+
+	// Project-specific memory count
+	if stats.ProjectObservations > 0 {
+		if useColors {
+			parts = append(parts, fmt.Sprintf("%sproject:%d memories%s", colorYellow, stats.ProjectObservations, reset))
+		} else {
+			parts = append(parts, fmt.Sprintf("project:%d memories", stats.ProjectObservations))
+		}
+	}
+
+	// Processing indicator
+	if stats.IsProcessing || stats.QueueDepth > 0 {
+		if useColors {
+			parts = append(parts, colorYellow+"processing..."+colorReset)
+		} else {
+			parts = append(parts, "processing...")
+		}
+	}
+
+	result := prefix + " " + indicator
+	for i, part := range parts {
+		if i == 0 {
+			result += " " + part
+		} else {
+			result += " | " + part
+		}
+	}
+
+	return result
+}
+
+// formatCompact returns a compact status line.
+func formatCompact(stats *WorkerStats, useColors bool) string {
+	// [m] ● 42/5/3 (28)
+	var prefix, indicator string
+	if useColors {
+		prefix = colorCyan + "[m]" + colorReset
+		indicator = colorGreen + "●" + colorReset
+	} else {
+		prefix = "[m]"
+		indicator = "●"
+	}
+
+	result := fmt.Sprintf("%s %s %d/%d/%d",
+		prefix, indicator,
+		stats.Retrieval.ObservationsServed,
+		stats.Retrieval.ContextInjections,
+		stats.Retrieval.SearchRequests,
+	)
+
+	if stats.ProjectObservations > 0 {
+		result += fmt.Sprintf(" (%d)", stats.ProjectObservations)
+	}
+
+	if stats.IsProcessing || stats.QueueDepth > 0 {
+		if useColors {
+			result += " " + colorYellow + "⚙" + colorReset
+		} else {
+			result += " ⚙"
+		}
+	}
+
+	return result
+}
+
+// formatMinimal returns a minimal status line.
+func formatMinimal(stats *WorkerStats, useColors bool) string {
+	// ● 42 obs
+	var indicator string
+	if useColors {
+		indicator = colorGreen + "●" + colorReset
+	} else {
+		indicator = "●"
+	}
+
+	result := fmt.Sprintf("%s %d", indicator, stats.Retrieval.ObservationsServed)
+
+	if stats.ProjectObservations > 0 {
+		result += fmt.Sprintf("/%d", stats.ProjectObservations)
+	}
+
+	return result
+}
+
+// formatOffline returns the offline status.
+func formatOffline() string {
+	return formatOfflineColored(true)
+}
+
+// formatOfflineColored returns the offline status with optional colors.
+func formatOfflineColored(useColors bool) string {
+	if useColors {
+		return colorCyan + "[mnemonic]" + colorReset + " " + colorGray + "○" + colorReset
+	}
+	return "[mnemonic] ○"
+}
+
+// formatStartingColored returns the starting status with optional colors.
+func formatStartingColored(useColors bool) string {
+	if useColors {
+		return colorCyan + "[mnemonic]" + colorReset + " " + colorYellow + "◐" + colorReset + " starting"
+	}
+	return "[mnemonic] ◐ starting"
+}
diff --git a/docs/src/App.vue b/docs/src/App.vue
index 1b27cc0..aa6122f 100644
--- a/docs/src/App.vue
+++ b/docs/src/App.vue
@@ -431,7 +431,7 @@ const features = [
   { icon: 'fas fa-folder-tree', title: 'Project-aware context', description: 'Your React knowledge stays in React projects. Your Go patterns stay in Go projects. No context pollution.' },
   { icon: 'fas fa-globe', title: 'Shared best practices', description: 'Security patterns, performance tips, and universal learnings automatically available across all your projects.' },
   { icon: 'fas fa-search', title: 'Finds what matters', description: 'Semantic search finds relevant memories even when you don\'t remember the exact words. "That auth thing" just works.' },
-  { icon: 'fas fa-credit-card', title: 'No API keys needed', description: 'Uses Claude Code CLI - works with your existing Claude Pro/Max subscription. No separate API costs.' },
+  { icon: 'fas fa-chart-line', title: 'Live statusline', description: 'Real-time metrics right in Claude Code: memories served, searches performed, and project memory count at a glance.' },
   { icon: 'fas fa-lock', title: '100% private', description: 'Your code context never leaves your machine. No telemetry. No cloud sync. Your memories are yours.' },
 ]
 
diff --git a/internal/db/sqlite/observation.go b/internal/db/sqlite/observation.go
index 46a0321..3b4eb32 100644
--- a/internal/db/sqlite/observation.go
+++ b/internal/db/sqlite/observation.go
@@ -74,12 +74,16 @@ func (s *ObservationStore) StoreObservation(ctx context.Context, sdkSessionID, p
 
 	id, _ := result.LastInsertId()
 
-	// Cleanup old observations beyond the limit for this project
+	// Cleanup old observations beyond the limit for this project (async to not block handler)
 	if project != "" {
-		deletedIDs, _ := s.CleanupOldObservations(ctx, project)
-		if len(deletedIDs) > 0 && s.cleanupFunc != nil {
-			s.cleanupFunc(ctx, deletedIDs)
-		}
+		go func(proj string) {
+			cleanupCtx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
+			defer cancel()
+			deletedIDs, _ := s.CleanupOldObservations(cleanupCtx, proj)
+			if len(deletedIDs) > 0 && s.cleanupFunc != nil {
+				s.cleanupFunc(cleanupCtx, deletedIDs)
+			}
+		}(project)
 	}
 
 	return id, nowEpoch, nil
diff --git a/internal/db/sqlite/prompt.go b/internal/db/sqlite/prompt.go
index e4a9b32..4bde673 100644
--- a/internal/db/sqlite/prompt.go
+++ b/internal/db/sqlite/prompt.go
@@ -51,11 +51,15 @@ func (s *PromptStore) SaveUserPromptWithMatches(ctx context.Context, claudeSessi
 
 	id, _ := result.LastInsertId()
 
-	// Cleanup old prompts beyond the global limit
-	deletedIDs, _ := s.CleanupOldPrompts(ctx)
-	if len(deletedIDs) > 0 && s.cleanupFunc != nil {
-		s.cleanupFunc(ctx, deletedIDs)
-	}
+	// Cleanup old prompts beyond the global limit (async to not block handler)
+	go func() {
+		cleanupCtx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
+		defer cancel()
+		deletedIDs, _ := s.CleanupOldPrompts(cleanupCtx)
+		if len(deletedIDs) > 0 && s.cleanupFunc != nil {
+			s.cleanupFunc(cleanupCtx, deletedIDs)
+		}
+	}()
 
 	return id, nil
 }
diff --git a/internal/worker/handlers.go b/internal/worker/handlers.go
index 95f55a2..9d195d2 100644
--- a/internal/worker/handlers.go
+++ b/internal/worker/handlers.go
@@ -514,7 +514,7 @@ func (s *Service) handleGetStats(w http.ResponseWriter, r *http.Request) {
 	retrievalStats := s.GetRetrievalStats()
 	sessionsToday, _ := s.sessionStore.GetSessionsToday(r.Context())
 
-	writeJSON(w, map[string]interface{}{
+	response := map[string]interface{}{
 		"uptime":           time.Since(s.startTime).String(),
 		"activeSessions":   s.sessionManager.GetActiveSessionCount(),
 		"queueDepth":       s.sessionManager.GetTotalQueueDepth(),
@@ -522,7 +522,19 @@ func (s *Service) handleGetStats(w http.ResponseWriter, r *http.Request) {
 		"connectedClients": s.sseBroadcaster.ClientCount(),
 		"sessionsToday":    sessionsToday,
 		"retrieval":        retrievalStats,
-	})
+		"ready":            s.ready.Load(),
+	}
+
+	// Include project-specific observation count if project is specified
+	if project := r.URL.Query().Get("project"); project != "" {
+		count, err := s.observationStore.GetObservationCount(r.Context(), project)
+		if err == nil {
+			response["projectObservations"] = count
+			response["project"] = project
+		}
+	}
+
+	writeJSON(w, response)
 }
 
 // handleGetRetrievalStats returns detailed retrieval statistics.
diff --git a/internal/worker/sse/broadcaster.go b/internal/worker/sse/broadcaster.go
index 1b2934e..b6e8d96 100644
--- a/internal/worker/sse/broadcaster.go
+++ b/internal/worker/sse/broadcaster.go
@@ -6,10 +6,17 @@ import (
 	"fmt"
 	"net/http"
 	"sync"
+	"time"
 
 	"github.com/rs/zerolog/log"
 )
 
+const (
+	// WriteTimeout is the timeout for writing to SSE clients.
+	// Prevents blocking on stale connections.
+	WriteTimeout = 2 * time.Second
+)
+
 // Client represents a connected SSE client.
 type Client struct {
 	ID      string
@@ -101,6 +108,7 @@ func (b *Broadcaster) removeClientByID(id string) {
 }
 
 // Broadcast sends a message to all connected clients.
+// Uses non-blocking writes with timeout to prevent stale connections from blocking.
 func (b *Broadcaster) Broadcast(data interface{}) {
 	jsonData, err := json.Marshal(data)
 	if err != nil {
@@ -117,30 +125,67 @@ func (b *Broadcaster) Broadcast(data interface{}) {
 	}
 	b.mu.RUnlock()
 
-	// Track dead clients for removal
-	var deadClients []*Client
+	if len(clients) == 0 {
+		return
+	}
+
+	// Use a channel to collect dead clients from concurrent writes
+	deadClientsCh := make(chan string, len(clients))
+	var wg sync.WaitGroup
 
 	for _, client := range clients {
 		select {
 		case <-client.Done:
 			continue
 		default:
-			_, err := client.Writer.Write([]byte(message))
-			if err != nil {
-				log.Debug().
-					Str("clientId", client.ID).
-					Err(err).
-					Msg("Failed to write to SSE client, marking for removal")
-				deadClients = append(deadClients, client)
-				continue
-			}
-			client.Flusher.Flush()
+			wg.Add(1)
+			go func(c *Client) {
+				defer wg.Done()
+				b.writeToClient(c, message, deadClientsCh)
+			}(client)
 		}
 	}
 
-	// Remove dead clients outside the iteration
-	for _, client := range deadClients {
-		b.removeClientByID(client.ID)
+	// Wait for all writes to complete (with their individual timeouts)
+	wg.Wait()
+	close(deadClientsCh)
+
+	// Remove dead clients
+	for clientID := range deadClientsCh {
+		b.removeClientByID(clientID)
+	}
+}
+
+// writeToClient writes a message to a single client with timeout.
+func (b *Broadcaster) writeToClient(client *Client, message string, deadCh chan<- string) {
+	// Use a timeout channel to prevent blocking on stale connections
+	done := make(chan struct{})
+
+	go func() {
+		defer close(done)
+		_, err := client.Writer.Write([]byte(message))
+		if err != nil {
+			log.Debug().
+				Str("clientId", client.ID).
+				Err(err).
+				Msg("Failed to write to SSE client, marking for removal")
+			deadCh <- client.ID
+			return
+		}
+		client.Flusher.Flush()
+	}()
+
+	select {
+	case <-done:
+		// Write completed successfully
+	case <-time.After(WriteTimeout):
+		log.Warn().
+			Str("clientId", client.ID).
+			Dur("timeout", WriteTimeout).
+			Msg("SSE write timed out, marking client for removal")
+		deadCh <- client.ID
+	case <-client.Done:
+		// Client disconnected during write
 	}
 }
 
diff --git a/plugin/.claude-plugin/marketplace.json.tpl b/plugin/.claude-plugin/marketplace.json.tpl
new file mode 100644
index 0000000..343eca7
--- /dev/null
+++ b/plugin/.claude-plugin/marketplace.json.tpl
@@ -0,0 +1,22 @@
+{
+  "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
+  "name": "claude-mnemonic",
+  "version": "{{ .Version }}",
+  "description": "Persistent memory system for Claude Code - stores observations, session summaries, and user prompts with semantic search",
+  "owner": {
+    "name": "lukaszraczylo",
+    "email": "lukaszraczylo@users.noreply.github.com"
+  },
+  "plugins": [
+    {
+      "name": "claude-mnemonic",
+      "description": "Persistent memory system for Claude Code - Go implementation with SQLite and ChromaDB",
+      "version": "{{ .Version }}",
+      "author": {
+        "name": "lukaszraczylo"
+      },
+      "source": "./",
+      "category": "productivity"
+    }
+  ]
+}
diff --git a/plugin/.claude-plugin/plugin.json.tpl b/plugin/.claude-plugin/plugin.json.tpl
new file mode 100644
index 0000000..b6124f9
--- /dev/null
+++ b/plugin/.claude-plugin/plugin.json.tpl
@@ -0,0 +1,9 @@
+{
+  "name": "claude-mnemonic",
+  "version": "{{ .Version }}",
+  "description": "Persistent memory system for Claude Code - Go implementation with SQLite and ChromaDB",
+  "author": {
+    "name": "lukaszraczylo",
+    "email": "lukaszraczylo@users.noreply.github.com"
+  }
+}
diff --git a/scripts/install.ps1 b/scripts/install.ps1
index fbd0eb7..73181ed 100644
--- a/scripts/install.ps1
+++ b/scripts/install.ps1
@@ -27,9 +27,28 @@ function Write-Error { param($Message) Write-Host "[ERROR] $Message" -Foreground
 
 function Get-LatestVersion {
     try {
-        $release = Invoke-RestMethod -Uri "https://api.github.com/repos/$GitHubRepo/releases/latest"
+        $headers = @{}
+        if ($env:GITHUB_TOKEN) {
+            $headers["Authorization"] = "token $env:GITHUB_TOKEN"
+        }
+        $release = Invoke-RestMethod -Uri "https://api.github.com/repos/$GitHubRepo/releases/latest" -Headers $headers
         return $release.tag_name
     } catch {
+        $errorMsg = $_.Exception.Message
+        if ($errorMsg -match "rate limit" -or $_.Exception.Response.StatusCode -eq 403) {
+            Write-Host ""
+            Write-Host "[ERROR] GitHub API rate limit exceeded." -ForegroundColor Red
+            Write-Host ""
+            Write-Host "You have a few options:" -ForegroundColor Yellow
+            Write-Host "  1. Wait ~1 hour for the rate limit to reset"
+            Write-Host "  2. Specify a version manually:"
+            Write-Host "     `$env:MNEMONIC_VERSION = 'v0.6.1'; irm https://raw.githubusercontent.com/$GitHubRepo/main/scripts/install.ps1 | iex" -ForegroundColor Cyan
+            Write-Host "  3. Use a GitHub token (set `$env:GITHUB_TOKEN)"
+            Write-Host "  4. Clone and build from source:"
+            Write-Host "     git clone https://github.com/$GitHubRepo.git" -ForegroundColor Cyan
+            Write-Host "     cd claude-mnemonic; make build; make install" -ForegroundColor Cyan
+            exit 1
+        }
         Write-Error "Failed to fetch latest version from GitHub: $_"
     }
 }
@@ -88,6 +107,14 @@ function Register-Plugin {
 
     # Ensure directories exist
     New-Item -ItemType Directory -Path "$env:USERPROFILE\.claude\plugins" -Force | Out-Null
+
+    # Clean up old cache versions to prevent stale binaries
+    $CacheBase = Split-Path $CachePath -Parent
+    if (Test-Path $CacheBase) {
+        Write-Info "Cleaning up old cache versions..."
+        Get-ChildItem -Path $CacheBase -Directory | Where-Object { $_.Name -ne $VersionClean } | Remove-Item -Recurse -Force -ErrorAction SilentlyContinue
+    }
+
     New-Item -ItemType Directory -Path $CachePath -Force | Out-Null
 
     # Create JSON files if they don't exist
@@ -132,8 +159,19 @@ function Register-Plugin {
             $Settings | Add-Member -NotePropertyName "enabledPlugins" -NotePropertyValue @{} -Force
         }
         $Settings.enabledPlugins | Add-Member -NotePropertyName $PluginKey -NotePropertyValue $true -Force
+
+        # Configure statusline
+        $StatuslineCmd = "$InstallDir\hooks\statusline.exe"
+        $StatuslineEntry = @{
+            type = "command"
+            command = $StatuslineCmd
+            padding = 0
+        }
+        $Settings | Add-Member -NotePropertyName "statusLine" -NotePropertyValue $StatuslineEntry -Force
+
         $Settings | ConvertTo-Json -Depth 10 | Out-File -Encoding UTF8 $SettingsFile
         Write-Success "Plugin enabled in settings.json"
+        Write-Success "Statusline configured in settings.json"
 
         # Update known_marketplaces.json
         $Marketplaces = Get-Content $MarketplacesFile -Raw | ConvertFrom-Json
@@ -194,8 +232,12 @@ function Uninstall-ClaudeMnemonic {
             $Settings = Get-Content $SettingsFile -Raw | ConvertFrom-Json
             if ($Settings.enabledPlugins) {
                 $Settings.enabledPlugins.PSObject.Properties.Remove($PluginKey)
-                $Settings | ConvertTo-Json -Depth 10 | Out-File -Encoding UTF8 $SettingsFile
             }
+            # Remove statusline if it's ours
+            if ($Settings.statusLine -and $Settings.statusLine.command -match "claude-mnemonic") {
+                $Settings.PSObject.Properties.Remove("statusLine")
+            }
+            $Settings | ConvertTo-Json -Depth 10 | Out-File -Encoding UTF8 $SettingsFile
         }
         if (Test-Path $MarketplacesFile) {
             $Marketplaces = Get-Content $MarketplacesFile -Raw | ConvertFrom-Json
diff --git a/scripts/install.sh b/scripts/install.sh
index 58010fc..95334ed 100755
--- a/scripts/install.sh
+++ b/scripts/install.sh
@@ -81,11 +81,44 @@ detect_platform() {
 
 # Get the latest release version from GitHub
 get_latest_version() {
-    local version
-    version=$(curl -sS "https://api.github.com/repos/${GITHUB_REPO}/releases/latest" | grep '"tag_name":' | sed -E 's/.*"([^"]+)".*/\1/')
+    local response version curl_opts
+
+    # Use GitHub token if available (higher rate limit)
+    curl_opts=(-sS)
+    if [[ -n "${GITHUB_TOKEN:-}" ]]; then
+        curl_opts+=(-H "Authorization: token ${GITHUB_TOKEN}")
+    fi
+
+    # Fetch with error handling
+    response=$(curl "${curl_opts[@]}" "https://api.github.com/repos/${GITHUB_REPO}/releases/latest" 2>&1)
+
+    # Check for rate limiting
+    if echo "$response" | grep -q "API rate limit exceeded"; then
+        echo ""
+        error "GitHub API rate limit exceeded.
+
+You have a few options:
+  1. Wait ~1 hour for the rate limit to reset
+  2. Specify a version manually:
+     curl -sSL https://raw.githubusercontent.com/${GITHUB_REPO}/main/scripts/install.sh | bash -s -- v0.6.1
+  3. Use a GitHub token (set GITHUB_TOKEN environment variable)
+  4. Clone and build from source:
+     git clone https://github.com/${GITHUB_REPO}.git
+     cd claude-mnemonic && make build && make install"
+    fi
+
+    # Check for other API errors
+    if echo "$response" | grep -q '"message":'; then
+        local msg
+        msg=$(echo "$response" | grep '"message":' | sed -E 's/.*"message": *"([^"]+)".*/\1/')
+        error "GitHub API error: $msg"
+    fi
+
+    # Extract version
+    version=$(echo "$response" | grep '"tag_name":' | sed -E 's/.*"([^"]+)".*/\1/')
 
     if [[ -z "$version" ]]; then
-        error "Failed to fetch latest version from GitHub"
+        error "Failed to fetch latest version from GitHub. Response: $response"
     fi
 
     echo "$version"
@@ -151,6 +184,15 @@ register_plugin() {
 
     # Ensure directories exist
     mkdir -p "$HOME/.claude/plugins"
+
+    # Clean up old cache versions to prevent stale binaries
+    local cache_base
+    cache_base=$(dirname "$CACHE_DIR")
+    if [[ -d "$cache_base" ]]; then
+        info "Cleaning up old cache versions..."
+        find "$cache_base" -mindepth 1 -maxdepth 1 -type d ! -name "${version#v}" -exec rm -rf {} \; 2>/dev/null || true
+    fi
+
     mkdir -p "${CACHE_DIR}/${version}"
 
     # Create JSON files if they don't exist
@@ -193,12 +235,24 @@ EOF
 
     success "Plugin registered in installed_plugins.json"
 
-    # Enable in settings.json
-    jq --arg key "$PLUGIN_KEY" \
-        '.enabledPlugins //= {} | .enabledPlugins[$key] = true' "$SETTINGS_FILE" > "${SETTINGS_FILE}.tmp" \
+    # Enable in settings.json and configure statusline
+    local statusline_cmd="$INSTALL_DIR/hooks/statusline"
+    local statusline_entry
+    statusline_entry=$(cat <<EOF
+{
+    "type": "command",
+    "command": "$statusline_cmd",
+    "padding": 0
+}
+EOF
+)
+
+    jq --arg key "$PLUGIN_KEY" --argjson statusline "$statusline_entry" \
+        '.enabledPlugins //= {} | .enabledPlugins[$key] = true | .statusLine = $statusline' "$SETTINGS_FILE" > "${SETTINGS_FILE}.tmp" \
         && mv "${SETTINGS_FILE}.tmp" "$SETTINGS_FILE"
 
     success "Plugin enabled in settings.json"
+    success "Statusline configured in settings.json"
 
     # Register marketplace
     local marketplace_entry
@@ -394,7 +448,8 @@ if [[ "${1:-}" == "--uninstall" ]]; then
             jq 'del(.plugins["'"$PLUGIN_KEY"'"])' "$PLUGINS_FILE" > "${PLUGINS_FILE}.tmp" && mv "${PLUGINS_FILE}.tmp" "$PLUGINS_FILE"
         fi
         if [[ -f "$SETTINGS_FILE" ]]; then
-            jq 'del(.enabledPlugins["'"$PLUGIN_KEY"'"])' "$SETTINGS_FILE" > "${SETTINGS_FILE}.tmp" && mv "${SETTINGS_FILE}.tmp" "$SETTINGS_FILE"
+            # Remove plugin from enabled plugins and remove statusline if it's ours
+            jq 'del(.enabledPlugins["'"$PLUGIN_KEY"'"]) | if .statusLine.command | test("claude-mnemonic") then del(.statusLine) else . end' "$SETTINGS_FILE" > "${SETTINGS_FILE}.tmp" && mv "${SETTINGS_FILE}.tmp" "$SETTINGS_FILE"
         fi
         if [[ -f "$MARKETPLACES_FILE" ]]; then
             jq 'del(.["claude-mnemonic"])' "$MARKETPLACES_FILE" > "${MARKETPLACES_FILE}.tmp" && mv "${MARKETPLACES_FILE}.tmp" "$MARKETPLACES_FILE"
diff --git a/scripts/register-plugin.sh b/scripts/register-plugin.sh
index bee859e..c0961da 100755
--- a/scripts/register-plugin.sh
+++ b/scripts/register-plugin.sh
@@ -9,13 +9,22 @@ MARKETPLACES_FILE="$HOME/.claude/plugins/known_marketplaces.json"
 PLUGIN_KEY="claude-mnemonic@claude-mnemonic"
 MARKETPLACE_NAME="claude-mnemonic"
 MARKETPLACE_PATH="$HOME/.claude/plugins/marketplaces/claude-mnemonic"
-CACHE_PATH="$HOME/.claude/plugins/cache/claude-mnemonic/claude-mnemonic/1.0.0"
-VERSION="1.0.0"
+
+# Get version from git tags (same as Makefile), or use argument if provided
+VERSION="${1:-$(git describe --tags --always --dirty 2>/dev/null || echo "dev")}"
+CACHE_BASE="$HOME/.claude/plugins/cache/claude-mnemonic/claude-mnemonic"
+CACHE_PATH="$CACHE_BASE/$VERSION"
 TIMESTAMP=$(date -u +"%Y-%m-%dT%H:%M:%S.000Z")
 
 # Ensure plugins directory exists
 mkdir -p "$HOME/.claude/plugins"
 
+# Clean up old cache versions to prevent stale binaries
+if [ -d "$CACHE_BASE" ]; then
+    echo "Cleaning up old cache versions..."
+    find "$CACHE_BASE" -mindepth 1 -maxdepth 1 -type d ! -name "$VERSION" -exec rm -rf {} \; 2>/dev/null || true
+fi
+
 # Create installed_plugins.json if it doesn't exist
 if [ ! -f "$PLUGINS_FILE" ]; then
     echo '{"version": 2, "plugins": {}}' > "$PLUGINS_FILE"
@@ -60,13 +69,24 @@ EOF
 
     echo "Plugin registered in installed_plugins.json"
 
-    # Enable the plugin in settings.json
+    # Enable the plugin in settings.json and configure statusline
     # First ensure enabledPlugins object exists, then add our plugin
-    jq --arg key "$PLUGIN_KEY" \
-        '.enabledPlugins //= {} | .enabledPlugins[$key] = true' "$SETTINGS_FILE" > "${SETTINGS_FILE}.tmp" \
+    STATUSLINE_CMD="$MARKETPLACE_PATH/hooks/statusline"
+    STATUSLINE_ENTRY=$(cat <<EOF
+{
+    "type": "command",
+    "command": "$STATUSLINE_CMD",
+    "padding": 0
+}
+EOF
+)
+
+    jq --arg key "$PLUGIN_KEY" --argjson statusline "$STATUSLINE_ENTRY" \
+        '.enabledPlugins //= {} | .enabledPlugins[$key] = true | .statusLine = $statusline' "$SETTINGS_FILE" > "${SETTINGS_FILE}.tmp" \
         && mv "${SETTINGS_FILE}.tmp" "$SETTINGS_FILE"
 
     echo "Plugin enabled in settings.json"
+    echo "Statusline configured in settings.json"
 
     # Register the marketplace in known_marketplaces.json
     MARKETPLACE_ENTRY=$(cat <<EOF