Fix Neo4j tag filter returning all events instead of filtering (v0.49.2)

- Change OPTIONAL MATCH to EXISTS subquery for tag filtering in Neo4j - OPTIONAL MATCH returned rows even when tags didn't match (NULL values) - EXISTS subquery correctly requires matching tags to exist - Strip "#" prefix from filter tag types before matching - Filters use "#d", "#p", "#e" but events store tags without prefix - Add trace-level logging for Neo4j query debugging - Add comprehensive tests for Neo4j query builder - Clean up temporary debug logging from handle-req.go Files modified: - pkg/neo4j/query-events.go: Fix tag filtering with EXISTS subquery - pkg/neo4j/query-events_test.go: Add query builder tests - app/handle-req.go: Remove debug logging - pkg/version/version: Bump to v0.49.2 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
feat: add NIP-99/Gamma Markets kind categories for curating mode
2026-01-14 17:49:46 +01:00 · 2026-01-14 17:16:57 +01:00 · 2026-01-14 10:56:31 +01:00 · 2026-01-14 08:09:39 +01:00 · 2026-01-11 11:29:31 +01:00 · 2026-01-11 11:14:20 +01:00
34 changed files with 1470 additions and 138 deletions
--- a/.claude/commands/release.md
+++ b/.claude/commands/release.md
@@ -49,10 +49,12 @@ If no argument provided, default to `patch`.
    GIT_SSH_COMMAND="ssh -i ~/.ssh/gitmlekudev" git push ssh://mleku@git.mleku.dev:2222/mleku/next.orly.dev.git main --tags
    ```

-11. **Deploy to VPS** by running:
-    ```
-    ssh relay.orly.dev 'cd ~/src/next.orly.dev && git stash && git pull origin main && export PATH=$PATH:~/go/bin && CGO_ENABLED=0 go build -o ~/.local/bin/next.orly.dev && sudo /usr/sbin/setcap cap_net_bind_service=+ep ~/.local/bin/next.orly.dev && sudo systemctl restart orly && ~/.local/bin/next.orly.dev version'
+11. **Deploy to relay.orly.dev** (ARM64):
+    Build on remote (faster than uploading cross-compiled binary due to slow local bandwidth):
+    ```bash
+    ssh relay.orly.dev 'cd ~/src/next.orly.dev && git pull origin main && GOPATH=$HOME CGO_ENABLED=0 ~/go/bin/go build -o ~/.local/bin/next.orly.dev && sudo /usr/sbin/setcap cap_net_bind_service=+ep ~/.local/bin/next.orly.dev && sudo systemctl restart orly && ~/.local/bin/next.orly.dev version'
    ```
+    Note: setcap must be re-applied after each binary rebuild to allow binding to ports 80/443.

 12. **Report completion** with the new version and commit hash

--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -40,7 +40,7 @@ NOSTR_SECRET_KEY=nsec1... ./nurl https://relay.example.com/api/logs/clear
 |----------|---------|-------------|
 | `ORLY_PORT` | 3334 | Server port |
 | `ORLY_LOG_LEVEL` | info | trace/debug/info/warn/error |
-| `ORLY_DB_TYPE` | badger | badger/neo4j/wasmdb |
+| `ORLY_DB_TYPE` | badger | badger/bbolt/neo4j/wasmdb |
 | `ORLY_POLICY_ENABLED` | false | Enable policy system |
 | `ORLY_ACL_MODE` | none | none/follows/managed |
 | `ORLY_TLS_DOMAINS` | | Let's Encrypt domains |
@@ -67,6 +67,7 @@ app/
  web/               → Svelte frontend (embedded via go:embed)
 pkg/
  database/          → Database interface + Badger implementation
+  bbolt/             → BBolt backend (HDD-optimized, B+tree)
  neo4j/             → Neo4j backend with WoT extensions
  wasmdb/            → WebAssembly IndexedDB backend
  protocol/          → Nostr protocol (ws/, auth/, publish/)
@@ -149,12 +150,59 @@ Before enabling auth-required on any deployment:

 | Backend | Use Case | Build |
 |---------|----------|-------|
-| **Badger** (default) | Single-instance, embedded | Standard |
+| **Badger** (default) | Single-instance, SSD, high performance | Standard |
+| **BBolt** | HDD-optimized, large archives, lower memory | `ORLY_DB_TYPE=bbolt` |
 | **Neo4j** | Social graph, WoT queries | `ORLY_DB_TYPE=neo4j` |
 | **WasmDB** | Browser/WebAssembly | `GOOS=js GOARCH=wasm` |

 All implement `pkg/database.Database` interface.

+### Scaling for Large Archives
+
+For archives with millions of events, consider:
+
+**Option 1: Tune Badger (SSD recommended)**
+```bash
+# Increase caches for larger working set (requires more RAM)
+ORLY_DB_BLOCK_CACHE_MB=2048      # 2GB block cache
+ORLY_DB_INDEX_CACHE_MB=1024      # 1GB index cache
+ORLY_SERIAL_CACHE_PUBKEYS=500000 # 500k pubkeys
+ORLY_SERIAL_CACHE_EVENT_IDS=2000000  # 2M event IDs
+
+# Higher compression to reduce disk IO
+ORLY_DB_ZSTD_LEVEL=9             # Best compression ratio
+
+# Enable storage GC with aggressive eviction
+ORLY_GC_ENABLED=true
+ORLY_GC_BATCH_SIZE=5000
+ORLY_MAX_STORAGE_BYTES=107374182400  # 100GB cap
+```
+
+**Option 2: Use BBolt for HDD/Low-Memory Deployments**
+```bash
+ORLY_DB_TYPE=bbolt
+
+# Tune for your HDD
+ORLY_BBOLT_BATCH_MAX_EVENTS=10000   # Larger batches for HDD
+ORLY_BBOLT_BATCH_MAX_MB=256          # 256MB batch buffer
+ORLY_BBOLT_FLUSH_TIMEOUT_SEC=60      # Longer flush interval
+ORLY_BBOLT_BLOOM_SIZE_MB=32          # Larger bloom filter
+ORLY_BBOLT_MMAP_SIZE_MB=16384        # 16GB mmap (scales with DB size)
+```
+
+**Migration Between Backends**
+```bash
+# Migrate from Badger to BBolt
+./orly migrate --from badger --to bbolt
+
+# Migrate with custom target path
+./orly migrate --from badger --to bbolt --target-path /mnt/hdd/orly-archive
+```
+
+**BBolt vs Badger Trade-offs:**
+- BBolt: Lower memory, HDD-friendly, simpler (B+tree), slower random reads
+- Badger: Higher memory, SSD-optimized (LSM), faster concurrent access
+
 ## Logging (lol.mleku.dev)

 ```go
@@ -221,7 +269,8 @@ if (isValidNsec(nsec)) { ... }

 ## Dependencies

- `github.com/dgraph-io/badger/v4` - Badger DB
+- `github.com/dgraph-io/badger/v4` - Badger DB (LSM, SSD-optimized)
+- `go.etcd.io/bbolt` - BBolt DB (B+tree, HDD-optimized)
 - `github.com/neo4j/neo4j-go-driver/v5` - Neo4j
 - `github.com/gorilla/websocket` - WebSocket
 - `github.com/ebitengine/purego` - CGO-free C loading
--- a/app/blossom.go
+++ b/app/blossom.go
@@ -20,8 +20,12 @@ func initializeBlossomServer(
 	blossomCfg := &blossom.Config{
 		BaseURL:          "", // Will be set dynamically per request
 		MaxBlobSize:      100 * 1024 * 1024, // 100MB default
-		AllowedMimeTypes: nil,                // Allow all MIME types by default
+		AllowedMimeTypes: nil,               // Allow all MIME types by default
 		RequireAuth:      cfg.AuthRequired || cfg.AuthToWrite,
+		// Rate limiting for non-followed users
+		RateLimitEnabled: cfg.BlossomRateLimitEnabled,
+		DailyLimitMB:     cfg.BlossomDailyLimitMB,
+		BurstLimitMB:     cfg.BlossomBurstLimitMB,
 	}

 	// Create blossom server with relay's ACL registry
@@ -31,7 +35,12 @@ func initializeBlossomServer(
 	// We'll need to modify the handler to inject the baseURL per request
 	// For now, we'll use a middleware approach

-	log.I.F("blossom server initialized with ACL mode: %s", cfg.ACLMode)
+	if cfg.BlossomRateLimitEnabled {
+		log.I.F("blossom server initialized with ACL mode: %s, rate limit: %dMB/day (burst: %dMB)",
+			cfg.ACLMode, cfg.BlossomDailyLimitMB, cfg.BlossomBurstLimitMB)
+	} else {
+		log.I.F("blossom server initialized with ACL mode: %s", cfg.ACLMode)
+	}
 	return bs, nil
 }

--- a/app/config/config.go
+++ b/app/config/config.go
@@ -41,9 +41,9 @@ type C struct {
 	EnableShutdown      bool          `env:"ORLY_ENABLE_SHUTDOWN" default:"false" usage:"if true, expose /shutdown on the health port to gracefully stop the process (for profiling)"`
 	LogLevel            string        `env:"ORLY_LOG_LEVEL" default:"info" usage:"relay log level: fatal error warn info debug trace"`
 	DBLogLevel          string        `env:"ORLY_DB_LOG_LEVEL" default:"info" usage:"database log level: fatal error warn info debug trace"`
-	DBBlockCacheMB      int           `env:"ORLY_DB_BLOCK_CACHE_MB" default:"512" usage:"Badger block cache size in MB (higher improves read hit ratio)"`
-	DBIndexCacheMB      int           `env:"ORLY_DB_INDEX_CACHE_MB" default:"256" usage:"Badger index cache size in MB (improves index lookup performance)"`
-	DBZSTDLevel         int           `env:"ORLY_DB_ZSTD_LEVEL" default:"1" usage:"Badger ZSTD compression level (1=fast/500MB/s, 3=default, 9=best ratio, 0=disable)"`
+	DBBlockCacheMB      int           `env:"ORLY_DB_BLOCK_CACHE_MB" default:"1024" usage:"Badger block cache size in MB (higher improves read hit ratio, increase for large archives)"`
+	DBIndexCacheMB      int           `env:"ORLY_DB_INDEX_CACHE_MB" default:"512" usage:"Badger index cache size in MB (improves index lookup performance, increase for large archives)"`
+	DBZSTDLevel         int           `env:"ORLY_DB_ZSTD_LEVEL" default:"3" usage:"Badger ZSTD compression level (1=fast/500MB/s, 3=balanced, 9=best ratio/slower, 0=disable)"`
 	LogToStdout         bool          `env:"ORLY_LOG_TO_STDOUT" default:"false" usage:"log to stdout instead of stderr"`
 	LogBufferSize       int           `env:"ORLY_LOG_BUFFER_SIZE" default:"10000" usage:"number of log entries to keep in memory for web UI viewing (0 disables)"`
 	Pprof               string        `env:"ORLY_PPROF" usage:"enable pprof in modes: cpu,memory,allocation,heap,block,goroutine,threadcreate,mutex"`
@@ -69,12 +69,18 @@ type C struct {

 	// Progressive throttle for follows ACL mode - allows non-followed users to write with increasing delay
 	FollowsThrottleEnabled  bool          `env:"ORLY_FOLLOWS_THROTTLE" default:"false" usage:"enable progressive delay for non-followed users in follows ACL mode"`
-	FollowsThrottlePerEvent time.Duration `env:"ORLY_FOLLOWS_THROTTLE_INCREMENT" default:"200ms" usage:"delay added per event for non-followed users"`
+	FollowsThrottlePerEvent time.Duration `env:"ORLY_FOLLOWS_THROTTLE_INCREMENT" default:"25ms" usage:"delay added per event for non-followed users"`
 	FollowsThrottleMaxDelay time.Duration `env:"ORLY_FOLLOWS_THROTTLE_MAX" default:"60s" usage:"maximum throttle delay cap"`

-	// Blossom blob storage service level settings
+	// Blossom blob storage service settings
+	BlossomEnabled       bool   `env:"ORLY_BLOSSOM_ENABLED" default:"true" usage:"enable Blossom blob storage server (only works with Badger backend)"`
 	BlossomServiceLevels string `env:"ORLY_BLOSSOM_SERVICE_LEVELS" usage:"comma-separated list of service levels in format: name:storage_mb_per_sat_per_month (e.g., basic:1,premium:10)"`

+	// Blossom upload rate limiting (for non-followed users)
+	BlossomRateLimitEnabled bool  `env:"ORLY_BLOSSOM_RATE_LIMIT" default:"false" usage:"enable upload rate limiting for non-followed users"`
+	BlossomDailyLimitMB     int64 `env:"ORLY_BLOSSOM_DAILY_LIMIT_MB" default:"10" usage:"daily upload limit in MB for non-followed users (EMA averaged)"`
+	BlossomBurstLimitMB     int64 `env:"ORLY_BLOSSOM_BURST_LIMIT_MB" default:"50" usage:"max burst upload in MB (bucket cap)"`
+
 	// Web UI and dev mode settings
 	WebDisableEmbedded bool   `env:"ORLY_WEB_DISABLE" default:"false" usage:"disable serving the embedded web UI; useful for hot-reload during development"`
 	WebDevProxyURL     string `env:"ORLY_WEB_DEV_PROXY_URL" usage:"when ORLY_WEB_DISABLE is true, reverse-proxy non-API paths to this dev server URL (e.g. http://localhost:5173)"`
@@ -124,9 +130,9 @@ type C struct {
 	Neo4jMaxTxRetrySeconds int `env:"ORLY_NEO4J_MAX_TX_RETRY_SEC" default:"30" usage:"max seconds for retryable transaction attempts"`
 	Neo4jQueryResultLimit  int `env:"ORLY_NEO4J_QUERY_RESULT_LIMIT" default:"10000" usage:"max results returned per query (prevents unbounded memory usage, 0=unlimited)"`

-	// Advanced database tuning
-	SerialCachePubkeys  int `env:"ORLY_SERIAL_CACHE_PUBKEYS" default:"100000" usage:"max pubkeys to cache for compact event storage (default: 100000, ~3.2MB memory)"`
-	SerialCacheEventIds int `env:"ORLY_SERIAL_CACHE_EVENT_IDS" default:"500000" usage:"max event IDs to cache for compact event storage (default: 500000, ~16MB memory)"`
+	// Advanced database tuning (increase for large archives to reduce cache misses)
+	SerialCachePubkeys  int `env:"ORLY_SERIAL_CACHE_PUBKEYS" default:"250000" usage:"max pubkeys to cache for compact event storage (~8MB memory, increase for large archives)"`
+	SerialCacheEventIds int `env:"ORLY_SERIAL_CACHE_EVENT_IDS" default:"1000000" usage:"max event IDs to cache for compact event storage (~32MB memory, increase for large archives)"`

 	// Connection concurrency control
 	MaxHandlersPerConnection int `env:"ORLY_MAX_HANDLERS_PER_CONN" default:"100" usage:"max concurrent message handlers per WebSocket connection (limits goroutine growth under load)"`
--- a/app/handle-cashu.go
+++ b/app/handle-cashu.go
@@ -124,6 +124,17 @@ func (s *Server) handleCashuKeysets(w http.ResponseWriter, r *http.Request) {

 // handleCashuInfo handles GET /cashu/info - returns mint information.
 func (s *Server) handleCashuInfo(w http.ResponseWriter, r *http.Request) {
+	// CORS headers for browser-based CAT support detection
+	w.Header().Set("Access-Control-Allow-Origin", "*")
+	w.Header().Set("Access-Control-Allow-Methods", "GET, OPTIONS")
+	w.Header().Set("Access-Control-Allow-Headers", "Content-Type, Accept")
+
+	// Handle preflight
+	if r.Method == http.MethodOptions {
+		w.WriteHeader(http.StatusNoContent)
+		return
+	}
+
 	if s.CashuIssuer == nil {
 		http.Error(w, "Cashu tokens not enabled", http.StatusNotImplemented)
 		return
--- a/app/handle-event.go
+++ b/app/handle-event.go
@@ -21,7 +21,7 @@ import (
 )

 func (l *Listener) HandleEvent(msg []byte) (err error) {
-	log.D.F("HandleEvent: START handling event: %s", msg)
+	log.I.F("HandleEvent: START handling event: %s", string(msg[:min(200, len(msg))]))

 	// 1. Raw JSON validation (before unmarshal) - use validation service
 	if result := l.eventValidator.ValidateRawJSON(msg); !result.Valid {
@@ -231,6 +231,11 @@ func (l *Listener) HandleEvent(msg []byte) (err error) {

 	// Authorization check (policy + ACL) - use authorization service
 	decision := l.eventAuthorizer.Authorize(env.E, l.authedPubkey.Load(), l.remote, env.E.Kind)
+	// Debug: log ephemeral event authorization
+	if env.E.Kind >= 20000 && env.E.Kind < 30000 {
+		log.I.F("ephemeral auth check: kind %d, allowed=%v, reason=%s",
+			env.E.Kind, decision.Allowed, decision.DenyReason)
+	}
 	if !decision.Allowed {
 		log.D.F("HandleEvent: authorization denied: %s (requireAuth=%v)", decision.DenyReason, decision.RequireAuth)
 		if decision.RequireAuth {
@@ -256,14 +261,17 @@ func (l *Listener) HandleEvent(msg []byte) (err error) {
 	log.I.F("HandleEvent: authorized with access level %s", decision.AccessLevel)

 	// Progressive throttle for follows ACL mode (delays non-followed users)
-	if delay := l.getFollowsThrottleDelay(env.E); delay > 0 {
-		log.D.F("HandleEvent: applying progressive throttle delay of %v for %0x from %s",
-			delay, env.E.Pubkey, l.remote)
-		select {
-		case <-l.ctx.Done():
-			return l.ctx.Err()
-		case <-time.After(delay):
-			// Delay completed, continue processing
+	// Skip throttle if a Cashu Access Token is present (authenticated via CAT)
+	if l.cashuToken == nil {
+		if delay := l.getFollowsThrottleDelay(env.E); delay > 0 {
+			log.D.F("HandleEvent: applying progressive throttle delay of %v for %0x from %s",
+				delay, env.E.Pubkey, l.remote)
+			select {
+			case <-l.ctx.Done():
+				return l.ctx.Err()
+			case <-time.After(delay):
+				// Delay completed, continue processing
+			}
 		}
 	}

--- a/app/handle-nip86-curating.go
+++ b/app/handle-nip86-curating.go
@@ -531,11 +531,23 @@ func GetKindCategoriesInfo() []map[string]interface{} {
 			"kinds":       []int{1063, 20, 21, 22},
 		},
 		{
-			"id":          "marketplace",
-			"name":        "Marketplace",
-			"description": "Product listings, stalls, auctions",
+			"id":          "marketplace_nip15",
+			"name":        "Marketplace (NIP-15)",
+			"description": "Legacy NIP-15 stalls and products",
 			"kinds":       []int{30017, 30018, 30019, 30020, 1021, 1022},
 		},
+		{
+			"id":          "marketplace_nip99",
+			"name":        "Marketplace (NIP-99/Gamma)",
+			"description": "NIP-99 classified listings, collections, shipping, reviews (Plebeian Market)",
+			"kinds":       []int{30402, 30403, 30405, 30406, 31555},
+		},
+		{
+			"id":          "order_communication",
+			"name":        "Order Communication",
+			"description": "Gamma Markets order messages and payment receipts",
+			"kinds":       []int{16, 17},
+		},
 		{
 			"id":          "groups_nip29",
 			"name":        "Group Messaging (NIP-29)",
--- a/app/handle-req.go
+++ b/app/handle-req.go
@@ -34,7 +34,6 @@ import (

 func (l *Listener) HandleReq(msg []byte) (err error) {
 	log.D.F("handling REQ: %s", msg)
-	log.T.F("HandleReq: START processing from %s", l.remote)
 	// var rem []byte
 	env := reqenvelope.New()
 	if _, err = env.Unmarshal(msg); chk.E(err) {
--- a/app/main.go
+++ b/app/main.go
@@ -435,7 +435,7 @@ func Run(

 	// Initialize Blossom blob storage server (only for Badger backend)
 	// MUST be done before UserInterface() which registers routes
-	if badgerDB, ok := db.(*database.D); ok {
+	if badgerDB, ok := db.(*database.D); ok && cfg.BlossomEnabled {
 		log.I.F("Badger backend detected, initializing Blossom server...")
 		if l.blossomServer, err = initializeBlossomServer(ctx, cfg, badgerDB); err != nil {
 			log.E.F("failed to initialize blossom server: %v", err)
@@ -445,6 +445,8 @@ func Run(
 		} else {
 			log.W.F("blossom server initialization returned nil without error")
 		}
+	} else if !cfg.BlossomEnabled {
+		log.I.F("Blossom server disabled via ORLY_BLOSSOM_ENABLED=false")
 	} else {
 		log.I.F("Non-Badger backend detected (type: %T), Blossom server not available", db)
 	}
--- a/app/publisher.go
+++ b/app/publisher.go
@@ -159,12 +159,26 @@ func (p *P) Deliver(ev *event.E) {
 		sub Subscription
 	}
 	var deliveries []delivery
+	// Debug: log ephemeral event delivery attempts
+	isEphemeral := ev.Kind >= 20000 && ev.Kind < 30000
+	if isEphemeral {
+		var tagInfo string
+		if ev.Tags != nil {
+			tagInfo = string(ev.Tags.Marshal(nil))
+		}
+		log.I.F("ephemeral event kind %d, id %0x, checking %d connections for matches, tags: %s",
+			ev.Kind, ev.ID[:8], len(p.Map), tagInfo)
+	}
 	for w, subs := range p.Map {
 		for id, subscriber := range subs {
 			if subscriber.Match(ev) {
 				deliveries = append(
 					deliveries, delivery{w: w, id: id, sub: subscriber},
 				)
+			} else if isEphemeral {
+				// Debug: log why ephemeral events don't match
+				log.I.F("ephemeral event kind %d did NOT match subscription %s (filters: %s)",
+					ev.Kind, id, string(subscriber.S.Marshal(nil)))
 			}
 		}
 	}
--- a/app/web/dist/bundle.js
+++ b/app/web/dist/bundle.js
--- a/app/web/dist/bundle.js.map
+++ b/app/web/dist/bundle.js.map
--- a/app/web/scripts/fetch-kinds.js
+++ b/app/web/scripts/fetch-kinds.js
@@ -6,25 +6,35 @@

 import { fileURLToPath } from 'url';
 import { dirname, join } from 'path';
-import { writeFileSync } from 'fs';
+import { writeFileSync, existsSync } from 'fs';

 const __filename = fileURLToPath(import.meta.url);
 const __dirname = dirname(__filename);

 const KINDS_URL = 'https://git.mleku.dev/mleku/nostr/raw/branch/main/encoders/kind/kinds.json';
+const OUTPUT_PATH = join(__dirname, '..', 'src', 'eventKinds.js');

 async function fetchKinds() {
  console.log(`Fetching kinds from ${KINDS_URL}...`);

-  const response = await fetch(KINDS_URL);
-  if (!response.ok) {
-    throw new Error(`Failed to fetch kinds.json: ${response.status} ${response.statusText}`);
+  try {
+    const response = await fetch(KINDS_URL, { timeout: 10000 });
+    if (!response.ok) {
+      throw new Error(`HTTP ${response.status} ${response.statusText}`);
+    }
+
+    const data = await response.json();
+    console.log(`Fetched ${Object.keys(data.kinds).length} kinds (version: ${data.version})`);
+    return data;
+  } catch (error) {
+    // Check if we have an existing eventKinds.js we can use
+    if (existsSync(OUTPUT_PATH)) {
+      console.warn(`Warning: Could not fetch kinds.json (${error.message})`);
+      console.log(`Using existing ${OUTPUT_PATH}`);
+      return null; // Signal to skip generation
+    }
+    throw new Error(`Failed to fetch kinds.json and no existing file: ${error.message}`);
  }
-
-  const data = await response.json();
-  console.log(`Fetched ${Object.keys(data.kinds).length} kinds (version: ${data.version})`);
-
-  return data;
 }

 function generateEventKinds(data) {
@@ -202,14 +212,18 @@ export const kindCategories = [
 async function main() {
  try {
    const data = await fetchKinds();
+
+    // If fetchKinds returned null, we're using the existing file
+    if (data === null) {
+      console.log('Skipping generation, using existing eventKinds.js');
+      return;
+    }
+
    const kinds = generateEventKinds(data);
    const js = generateJS(kinds, data);

-    // Write to src/eventKinds.js
-    const outPath = join(__dirname, '..', 'src', 'eventKinds.js');
-
-    writeFileSync(outPath, js);
-    console.log(`Generated ${outPath} with ${kinds.length} kinds`);
+    writeFileSync(OUTPUT_PATH, js);
+    console.log(`Generated ${OUTPUT_PATH} with ${kinds.length} kinds`);
  } catch (error) {
    console.error('Error:', error.message);
    process.exit(1);
--- a/app/web/src/kindCategories.js
+++ b/app/web/src/kindCategories.js
@@ -30,11 +30,23 @@ export const curationKindCategories = [
    kinds: [1063, 20, 21, 22],
  },
  {
-    id: "marketplace",
-    name: "Marketplace",
-    description: "Product listings, stalls, and marketplace events",
+    id: "marketplace_nip15",
+    name: "Marketplace (NIP-15)",
+    description: "Legacy NIP-15 stalls and products",
    kinds: [30017, 30018, 30019, 30020],
  },
+  {
+    id: "marketplace_nip99",
+    name: "Marketplace (NIP-99/Gamma)",
+    description: "NIP-99 classified listings, collections, shipping, reviews (Plebeian Market)",
+    kinds: [30402, 30403, 30405, 30406, 31555],
+  },
+  {
+    id: "order_communication",
+    name: "Order Communication",
+    description: "Gamma Markets order messages and payment receipts (kinds 16, 17)",
+    kinds: [16, 17],
+  },
  {
    id: "groups_nip29",
    name: "Group Messaging (NIP-29)",
--- a/app/web/src/nostr.js
+++ b/app/web/src/nostr.js
@@ -179,6 +179,28 @@ export class Nip07Signer {
  }
 }

+// Merge two event arrays, deduplicating by event id
+// Newer events (by created_at) take precedence for same id
+function mergeAndDeduplicateEvents(cached, relay) {
+  const eventMap = new Map();
+
+  // Add cached events first
+  for (const event of cached) {
+    eventMap.set(event.id, event);
+  }
+
+  // Add/update with relay events (they may be newer)
+  for (const event of relay) {
+    const existing = eventMap.get(event.id);
+    if (!existing || event.created_at >= existing.created_at) {
+      eventMap.set(event.id, event);
+    }
+  }
+
+  // Return sorted by created_at descending (newest first)
+  return Array.from(eventMap.values()).sort((a, b) => b.created_at - a.created_at);
+}
+
 // IndexedDB helpers for unified event storage
 // This provides a local cache that all components can access
 const DB_NAME = "nostrCache";
@@ -573,9 +595,10 @@ export async function fetchEvents(filters, options = {}) {
  } = options;

  // Try to get cached events first if requested
+  let cachedEvents = [];
  if (useCache) {
    try {
-      const cachedEvents = await queryEventsFromDB(filters);
+      cachedEvents = await queryEventsFromDB(filters);
      if (cachedEvents.length > 0) {
        console.log(`Found ${cachedEvents.length} cached events in IndexedDB`);
      }
@@ -585,17 +608,19 @@ export async function fetchEvents(filters, options = {}) {
  }

  return new Promise((resolve, reject) => {
-    const events = [];
+    const relayEvents = [];
    const timeoutId = setTimeout(() => {
-      console.log(`Timeout reached after ${timeout}ms, returning ${events.length} events`);
+      console.log(`Timeout reached after ${timeout}ms, returning ${relayEvents.length} relay events`);
      sub.close();
-      
+
      // Store all received events in IndexedDB before resolving
-      if (events.length > 0) {
-        putEvents(events).catch(e => console.warn("Failed to cache events", e));
+      if (relayEvents.length > 0) {
+        putEvents(relayEvents).catch(e => console.warn("Failed to cache events", e));
      }
-      
-      resolve(events);
+
+      // Merge cached events with relay events, deduplicate by id
+      const mergedEvents = mergeAndDeduplicateEvents(cachedEvents, relayEvents);
+      resolve(mergedEvents);
    }, timeout);

    try {
@@ -615,22 +640,25 @@ export async function fetchEvents(filters, options = {}) {
              created_at: event.created_at,
              content_preview: event.content?.substring(0, 50)
            });
-            events.push(event);
-            
+            relayEvents.push(event);
+
            // Store event immediately in IndexedDB
            putEvent(event).catch(e => console.warn("Failed to cache event", e));
          },
          oneose() {
-            console.log(`✅ EOSE received for REQ [${subId}], got ${events.length} events`);
+            console.log(`✅ EOSE received for REQ [${subId}], got ${relayEvents.length} relay events`);
            clearTimeout(timeoutId);
            sub.close();
-            
+
            // Store all events in IndexedDB before resolving
-            if (events.length > 0) {
-              putEvents(events).catch(e => console.warn("Failed to cache events", e));
+            if (relayEvents.length > 0) {
+              putEvents(relayEvents).catch(e => console.warn("Failed to cache events", e));
            }
-            
-            resolve(events);
+
+            // Merge cached events with relay events, deduplicate by id
+            const mergedEvents = mergeAndDeduplicateEvents(cachedEvents, relayEvents);
+            console.log(`Merged ${cachedEvents.length} cached + ${relayEvents.length} relay = ${mergedEvents.length} total events`);
+            resolve(mergedEvents);
          }
        }
      );
--- a/docs/NIP-NRC.md
+++ b/docs/NIP-NRC.md
@@ -137,7 +137,7 @@ Where `payload` is the standard Nostr message array, e.g.:
 The encrypted content structure:
 ```json
 {
-  "type": "EVENT" | "OK" | "EOSE" | "NOTICE" | "CLOSED" | "COUNT" | "AUTH",
+  "type": "EVENT" | "OK" | "EOSE" | "NOTICE" | "CLOSED" | "COUNT" | "AUTH" | "CHUNK",
  "payload": <standard_nostr_response_array>
 }
 ```
@@ -150,6 +150,7 @@ Where `payload` is the standard Nostr response array, e.g.:
 - `["CLOSED", "<sub_id>", "<message>"]`
 - `["COUNT", "<sub_id>", {"count": <n>}]`
 - `["AUTH", "<challenge>"]`
+- `[<chunk_object>]` (for CHUNK type, see Message Segmentation)

 ### Session Management

@@ -168,6 +169,85 @@ The conversation key is derived from:
 - **Secret-based auth**: ECDH between client's secret key (derived from URI secret) and relay's public key
 - **CAT auth**: ECDH between client's Nostr key and relay's public key

+### Message Segmentation
+
+Some Nostr events exceed the typical relay message size limits (commonly 64KB). NRC supports message segmentation to handle large payloads by splitting them into multiple chunks.
+
+#### When to Chunk
+
+Senders SHOULD chunk messages when the JSON-serialized response exceeds 40KB. This threshold accounts for:
+- NIP-44 encryption overhead (~100 bytes)
+- Base64 encoding expansion (~33%)
+- Event wrapper overhead (tags, signature, etc.)
+
+#### Chunk Message Format
+
+When a response is too large, it is split into multiple CHUNK responses:
+
+```json
+{
+  "type": "CHUNK",
+  "payload": [{
+    "type": "CHUNK",
+    "messageId": "<uuid>",
+    "index": 0,
+    "total": 3,
+    "data": "<base64_encoded_chunk>"
+  }]
+}
+```
+
+Fields:
+- `messageId`: A unique identifier (UUID) for the chunked message, used to correlate chunks
+- `index`: Zero-based chunk index (0, 1, 2, ...)
+- `total`: Total number of chunks in this message
+- `data`: Base64-encoded segment of the original message
+
+#### Chunking Process (Sender)
+
+1. Serialize the original response message to JSON
+2. If the serialized length exceeds the threshold (40KB), proceed with chunking
+3. Encode the JSON string as UTF-8, then Base64 encode it
+4. Split the Base64 string into chunks of the maximum chunk size
+5. Generate a unique `messageId` (UUID recommended)
+6. Send each chunk as a separate CHUNK response event
+
+Example encoding (JavaScript):
+```javascript
+const encoded = btoa(unescape(encodeURIComponent(jsonString)))
+```
+
+#### Reassembly Process (Receiver)
+
+1. When receiving a CHUNK response, buffer it by `messageId`
+2. Track received chunks by `index`
+3. When all chunks are received (`chunks.size === total`):
+   a. Concatenate chunk data in index order (0, 1, 2, ...)
+   b. Base64 decode the concatenated string
+   c. Parse as UTF-8 JSON to recover the original response
+4. Process the reassembled response as normal
+5. Clean up the chunk buffer
+
+Example decoding (JavaScript):
+```javascript
+const jsonString = decodeURIComponent(escape(atob(concatenatedBase64)))
+const response = JSON.parse(jsonString)
+```
+
+#### Chunk Buffer Management
+
+Receivers MUST implement chunk buffer cleanup:
+- Discard incomplete chunk buffers after 60 seconds of inactivity
+- Limit the number of concurrent incomplete messages to prevent memory exhaustion
+- Log warnings when discarding stale buffers for debugging
+
+#### Ordering and Reliability
+
+- Chunks MAY arrive out of order; receivers MUST reassemble by index
+- Missing chunks result in message loss; the incomplete buffer is eventually discarded
+- Duplicate chunks (same messageId + index) SHOULD be ignored
+- Each chunk is sent as a separate encrypted NRC response event
+
 ### Authentication

 #### Secret-Based Authentication
@@ -208,6 +288,9 @@ The conversation key is derived from:
 4. Match responses using the `e` tag (references request event ID)
 5. Handle EOSE by waiting for kind 24892 with type "EOSE" in content
 6. For subscriptions, maintain mapping of internal sub IDs to tunnel session
+7. **Chunking**: Maintain a chunk buffer map keyed by `messageId`
+8. **Chunking**: When receiving CHUNK responses, buffer chunks and reassemble when complete
+9. **Chunking**: Implement 60-second timeout for incomplete chunk buffers

 ## Bridge Implementation Notes

@@ -217,10 +300,14 @@ The conversation key is derived from:
 4. Capture all relay responses and wrap in kind 24892
 5. Sign with relay's key and publish to rendezvous relay
 6. Maintain session state for subscription mapping
+7. **Chunking**: Check response size before sending; chunk if > 40KB
+8. **Chunking**: Use consistent messageId (UUID) across all chunks of a message
+9. **Chunking**: Send chunks in order (index 0, 1, 2, ...) for optimal reassembly

 ## Reference Implementations

- ORLY Relay: [https://git.mleku.dev/mleku/next.orly.dev](https://git.mleku.dev/mleku/next.orly.dev)
+- ORLY Relay (Bridge): [https://git.mleku.dev/mleku/next.orly.dev](https://git.mleku.dev/mleku/next.orly.dev)
+- Smesh Client: [https://git.mleku.dev/mleku/smesh](https://git.mleku.dev/mleku/smesh)

 ## See Also

--- a/go.mod
+++ b/go.mod
@@ -3,12 +3,14 @@ module next.orly.dev
 go 1.25.3

 require (
-	git.mleku.dev/mleku/nostr v1.0.12
+	git.mleku.dev/mleku/nostr v1.0.13
 	github.com/adrg/xdg v0.5.3
 	github.com/alexflint/go-arg v1.6.1
 	github.com/aperturerobotics/go-indexeddb v0.2.3
+	github.com/bits-and-blooms/bloom/v3 v3.7.1
 	github.com/decred/dcrd/dcrec/secp256k1/v4 v4.4.0
 	github.com/dgraph-io/badger/v4 v4.8.0
+	github.com/google/uuid v1.6.0
 	github.com/gorilla/websocket v1.5.3
 	github.com/hack-pad/safejs v0.1.1
 	github.com/kardianos/osext v0.0.0-20190222173326-2bc1f35cddc0
@@ -22,6 +24,7 @@ require (
 	github.com/stretchr/testify v1.11.1
 	github.com/vertex-lab/nostr-sqlite v0.3.2
 	go-simpler.org/env v0.12.0
+	go.etcd.io/bbolt v1.4.3
 	go.uber.org/atomic v1.11.0
 	golang.org/x/crypto v0.46.0
 	golang.org/x/lint v0.0.0-20241112194109-818c5a804067
@@ -37,7 +40,6 @@ require (
 	github.com/ImVexed/fasturl v0.0.0-20230304231329-4e41488060f3 // indirect
 	github.com/alexflint/go-scalar v1.2.0 // indirect
 	github.com/bits-and-blooms/bitset v1.24.2 // indirect
-	github.com/bits-and-blooms/bloom/v3 v3.7.1 // indirect
 	github.com/btcsuite/btcd/btcec/v2 v2.3.4 // indirect
 	github.com/btcsuite/btcd/chaincfg/chainhash v1.1.0 // indirect
 	github.com/bytedance/sonic v1.13.1 // indirect
@@ -56,7 +58,6 @@ require (
 	github.com/google/btree v1.1.2 // indirect
 	github.com/google/flatbuffers v25.9.23+incompatible // indirect
 	github.com/google/pprof v0.0.0-20251007162407-5df77e3f7d1d // indirect
-	github.com/google/uuid v1.6.0 // indirect
 	github.com/josharian/intern v1.0.0 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect
 	github.com/klauspost/cpuid/v2 v2.3.0 // indirect
@@ -72,7 +73,6 @@ require (
 	github.com/tidwall/match v1.1.1 // indirect
 	github.com/tidwall/pretty v1.2.1 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
-	go.etcd.io/bbolt v1.4.3 // indirect
 	go.opentelemetry.io/auto/sdk v1.2.1 // indirect
 	go.opentelemetry.io/otel v1.38.0 // indirect
 	go.opentelemetry.io/otel/metric v1.38.0 // indirect
--- a/go.sum
+++ b/go.sum
@@ -1,5 +1,5 @@
-git.mleku.dev/mleku/nostr v1.0.12 h1:bjsFUh1Q3fGpU7qsqxggGgrGGUt2OBdu1w8hjDM4gJE=
-git.mleku.dev/mleku/nostr v1.0.12/go.mod h1:kJwSMmLRnAJ7QJtgXDv2wGgceFU0luwVqrgAL3MI93M=
+git.mleku.dev/mleku/nostr v1.0.13 h1:FqeOQ9ZX8AFVsAI6XisQkB6cgmhn9DNQ2a8li9gx7aY=
+git.mleku.dev/mleku/nostr v1.0.13/go.mod h1:kJwSMmLRnAJ7QJtgXDv2wGgceFU0luwVqrgAL3MI93M=
 github.com/BurntSushi/toml v1.5.0 h1:W5quZX/G/csjUnuI8SUYlsHs9M38FC7znL0lIO+DvMg=
 github.com/BurntSushi/toml v1.5.0/go.mod h1:ukJfTF/6rtPPRCnwkur4qwRxa8vTRFBF0uk2lLoLwho=
 github.com/ImVexed/fasturl v0.0.0-20230304231329-4e41488060f3 h1:ClzzXMDDuUbWfNNZqGeYq4PnYOlwlOVIvSyNaIy0ykg=
@@ -161,6 +161,7 @@ github.com/tidwall/pretty v1.2.1 h1:qjsOFOWWQl+N3RsoF5/ssm1pHmJJwhjlSbZ51I6wMl4=
 github.com/tidwall/pretty v1.2.1/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
 github.com/twitchyliquid64/golang-asm v0.15.1 h1:SU5vSMR7hnwNxj24w34ZyCi/FmDZTkS4MhqMhdFk5YI=
 github.com/twitchyliquid64/golang-asm v0.15.1/go.mod h1:a1lVb/DtPvCB8fslRZhAngC2+aY1QWCk3Cedj/Gdt08=
+github.com/twmb/murmur3 v1.1.8 h1:8Yt9taO/WN3l08xErzjeschgZU2QSrwm1kclYq+0aRg=
 github.com/twmb/murmur3 v1.1.8/go.mod h1:Qq/R7NUyOfr65zD+6Q5IHKsJLwP7exErjN6lyyq3OSQ=
 github.com/vertex-lab/nostr-sqlite v0.3.2 h1:8nZYYIwiKnWLA446qA/wL/Gy+bU0kuaxdLfUyfeTt/E=
 github.com/vertex-lab/nostr-sqlite v0.3.2/go.mod h1:5bw1wMgJhSdrumsZAWxqy+P0u1g+q02PnlGQn15dnSM=
--- a/main.go
+++ b/main.go
@@ -24,7 +24,7 @@ import (
 	"next.orly.dev/pkg/acl"
 	"git.mleku.dev/mleku/nostr/crypto/keys"
 	"git.mleku.dev/mleku/nostr/encoders/bech32encoding"
-	_ "next.orly.dev/pkg/bbolt"   // Import for bbolt factory registration
+	bboltdb "next.orly.dev/pkg/bbolt" // Import for bbolt factory and type
 	"next.orly.dev/pkg/database"
 	neo4jdb "next.orly.dev/pkg/neo4j" // Import for neo4j factory and type
 	"git.mleku.dev/mleku/nostr/encoders/hex"
@@ -617,6 +617,10 @@ func main() {
 				n4jDB.MaxConcurrentQueries(),
 			)
 			log.I.F("rate limiter configured for Neo4j backend (target: %dMB)", targetMB)
+		} else if _, ok := db.(*bboltdb.B); ok {
+			// BBolt uses memory-mapped IO, so memory-only limiter is appropriate
+			limiter = ratelimit.NewMemoryOnlyLimiter(rlConfig)
+			log.I.F("rate limiter configured for BBolt backend (target: %dMB)", targetMB)
 		} else {
 			// For other backends, create a disabled limiter
 			limiter = ratelimit.NewDisabledLimiter()
--- a/pkg/acl/follows.go
+++ b/pkg/acl/follows.go
@@ -138,7 +138,7 @@ func (f *Follows) Configure(cfg ...any) (err error) {
 	if f.cfg.FollowsThrottleEnabled {
 		perEvent := f.cfg.FollowsThrottlePerEvent
 		if perEvent == 0 {
-			perEvent = 200 * time.Millisecond
+			perEvent = 25 * time.Millisecond
 		}
 		maxDelay := f.cfg.FollowsThrottleMaxDelay
 		if maxDelay == 0 {
--- a/pkg/blossom/handlers.go
+++ b/pkg/blossom/handlers.go
@@ -200,6 +200,12 @@ func (s *Server) handleUpload(w http.ResponseWriter, r *http.Request) {
 		return
 	}

+	// Check bandwidth rate limit (non-followed users)
+	if !s.checkBandwidthLimit(pubkey, remoteAddr, int64(len(body))) {
+		s.setErrorResponse(w, http.StatusTooManyRequests, "upload rate limit exceeded, try again later")
+		return
+	}
+
 	// Calculate SHA256 after auth check
 	sha256Hash := CalculateSHA256(body)
 	sha256Hex := hex.Enc(sha256Hash)
@@ -647,6 +653,12 @@ func (s *Server) handleMirror(w http.ResponseWriter, r *http.Request) {
 		return
 	}

+	// Check bandwidth rate limit (non-followed users)
+	if !s.checkBandwidthLimit(pubkey, remoteAddr, int64(len(body))) {
+		s.setErrorResponse(w, http.StatusTooManyRequests, "upload rate limit exceeded, try again later")
+		return
+	}
+
 	// Note: pubkey may be nil for anonymous uploads if ACL allows it

 	// Detect MIME type from remote response
@@ -726,6 +738,12 @@ func (s *Server) handleMediaUpload(w http.ResponseWriter, r *http.Request) {
 		return
 	}

+	// Check bandwidth rate limit (non-followed users)
+	if !s.checkBandwidthLimit(pubkey, remoteAddr, int64(len(body))) {
+		s.setErrorResponse(w, http.StatusTooManyRequests, "upload rate limit exceeded, try again later")
+		return
+	}
+
 	// Note: pubkey may be nil for anonymous uploads if ACL allows it

 	// Optimize media (placeholder - actual optimization would be implemented here)
--- a/pkg/blossom/ratelimit.go
+++ b/pkg/blossom/ratelimit.go
@@ -0,0 +1,131 @@
+package blossom
+
+import (
+	"sync"
+	"time"
+)
+
+// BandwidthState tracks upload bandwidth for an identity
+type BandwidthState struct {
+	BucketBytes   int64     // Current token bucket level (bytes available)
+	LastUpdate    time.Time // Last time bucket was updated
+}
+
+// BandwidthLimiter implements token bucket rate limiting for uploads.
+// Each identity gets a bucket that replenishes at dailyLimit/day rate.
+// Uploads consume tokens from the bucket.
+type BandwidthLimiter struct {
+	mu           sync.Mutex
+	states       map[string]*BandwidthState // keyed by pubkey hex or IP
+	dailyLimit   int64                      // bytes per day
+	burstLimit   int64                      // max bucket size (burst capacity)
+	refillRate   float64                    // bytes per second refill rate
+}
+
+// NewBandwidthLimiter creates a new bandwidth limiter.
+// dailyLimitMB is the average daily limit in megabytes.
+// burstLimitMB is the maximum burst capacity in megabytes.
+func NewBandwidthLimiter(dailyLimitMB, burstLimitMB int64) *BandwidthLimiter {
+	dailyBytes := dailyLimitMB * 1024 * 1024
+	burstBytes := burstLimitMB * 1024 * 1024
+
+	return &BandwidthLimiter{
+		states:     make(map[string]*BandwidthState),
+		dailyLimit: dailyBytes,
+		burstLimit: burstBytes,
+		refillRate: float64(dailyBytes) / 86400.0, // bytes per second
+	}
+}
+
+// CheckAndConsume checks if an upload of the given size is allowed for the identity,
+// and if so, consumes the tokens. Returns true if allowed, false if rate limited.
+// The identity should be pubkey hex for authenticated users, or IP for anonymous.
+func (bl *BandwidthLimiter) CheckAndConsume(identity string, sizeBytes int64) bool {
+	bl.mu.Lock()
+	defer bl.mu.Unlock()
+
+	now := time.Now()
+	state, exists := bl.states[identity]
+
+	if !exists {
+		// New identity starts with full burst capacity
+		state = &BandwidthState{
+			BucketBytes: bl.burstLimit,
+			LastUpdate:  now,
+		}
+		bl.states[identity] = state
+	} else {
+		// Refill bucket based on elapsed time
+		elapsed := now.Sub(state.LastUpdate).Seconds()
+		refill := int64(elapsed * bl.refillRate)
+		state.BucketBytes += refill
+		if state.BucketBytes > bl.burstLimit {
+			state.BucketBytes = bl.burstLimit
+		}
+		state.LastUpdate = now
+	}
+
+	// Check if upload fits in bucket
+	if state.BucketBytes >= sizeBytes {
+		state.BucketBytes -= sizeBytes
+		return true
+	}
+
+	return false
+}
+
+// GetAvailable returns the currently available bytes for an identity.
+func (bl *BandwidthLimiter) GetAvailable(identity string) int64 {
+	bl.mu.Lock()
+	defer bl.mu.Unlock()
+
+	state, exists := bl.states[identity]
+	if !exists {
+		return bl.burstLimit // New users have full capacity
+	}
+
+	// Calculate current level with refill
+	now := time.Now()
+	elapsed := now.Sub(state.LastUpdate).Seconds()
+	refill := int64(elapsed * bl.refillRate)
+	available := state.BucketBytes + refill
+	if available > bl.burstLimit {
+		available = bl.burstLimit
+	}
+
+	return available
+}
+
+// GetTimeUntilAvailable returns how long until the given bytes will be available.
+func (bl *BandwidthLimiter) GetTimeUntilAvailable(identity string, sizeBytes int64) time.Duration {
+	available := bl.GetAvailable(identity)
+	if available >= sizeBytes {
+		return 0
+	}
+
+	needed := sizeBytes - available
+	seconds := float64(needed) / bl.refillRate
+	return time.Duration(seconds * float64(time.Second))
+}
+
+// Cleanup removes entries that have fully replenished (at burst limit).
+func (bl *BandwidthLimiter) Cleanup() {
+	bl.mu.Lock()
+	defer bl.mu.Unlock()
+
+	now := time.Now()
+	for key, state := range bl.states {
+		elapsed := now.Sub(state.LastUpdate).Seconds()
+		refill := int64(elapsed * bl.refillRate)
+		if state.BucketBytes+refill >= bl.burstLimit {
+			delete(bl.states, key)
+		}
+	}
+}
+
+// Stats returns the number of tracked identities.
+func (bl *BandwidthLimiter) Stats() int {
+	bl.mu.Lock()
+	defer bl.mu.Unlock()
+	return len(bl.states)
+}
--- a/pkg/blossom/server.go
+++ b/pkg/blossom/server.go
@@ -19,6 +19,9 @@ type Server struct {
 	maxBlobSize      int64
 	allowedMimeTypes map[string]bool
 	requireAuth      bool
+
+	// Rate limiting for uploads
+	bandwidthLimiter *BandwidthLimiter
 }

 // Config holds configuration for the Blossom server
@@ -27,6 +30,11 @@ type Config struct {
 	MaxBlobSize      int64
 	AllowedMimeTypes []string
 	RequireAuth      bool
+
+	// Rate limiting (for non-followed users)
+	RateLimitEnabled bool
+	DailyLimitMB     int64
+	BurstLimitMB     int64
 }

 // NewServer creates a new Blossom server instance
@@ -48,6 +56,20 @@ func NewServer(db *database.D, aclRegistry *acl.S, cfg *Config) *Server {
 		}
 	}

+	// Initialize bandwidth limiter if enabled
+	var bwLimiter *BandwidthLimiter
+	if cfg.RateLimitEnabled {
+		dailyMB := cfg.DailyLimitMB
+		if dailyMB <= 0 {
+			dailyMB = 10 // 10MB default
+		}
+		burstMB := cfg.BurstLimitMB
+		if burstMB <= 0 {
+			burstMB = 50 // 50MB default burst
+		}
+		bwLimiter = NewBandwidthLimiter(dailyMB, burstMB)
+	}
+
 	return &Server{
 		db:               db,
 		storage:          storage,
@@ -56,6 +78,7 @@ func NewServer(db *database.D, aclRegistry *acl.S, cfg *Config) *Server {
 		maxBlobSize:      cfg.MaxBlobSize,
 		allowedMimeTypes: allowedMap,
 		requireAuth:      cfg.RequireAuth,
+		bandwidthLimiter: bwLimiter,
 	}
 }

@@ -208,6 +231,44 @@ func (s *Server) checkACL(
 	return actual >= required
 }

+// isRateLimitExempt returns true if the user is exempt from rate limiting.
+// Users with write access or higher (followed users, admins, owners) are exempt.
+func (s *Server) isRateLimitExempt(pubkey []byte, remoteAddr string) bool {
+	if s.acl == nil {
+		return true // No ACL configured, no rate limiting
+	}
+
+	level := s.acl.GetAccessLevel(pubkey, remoteAddr)
+
+	// Followed users get "write" level, admins/owners get higher
+	// Only "read" and "none" are rate limited
+	return level == "write" || level == "admin" || level == "owner"
+}
+
+// checkBandwidthLimit checks if the upload is allowed under rate limits.
+// Returns true if allowed, false if rate limited.
+// Exempt users (followed, admin, owner) always return true.
+func (s *Server) checkBandwidthLimit(pubkey []byte, remoteAddr string, sizeBytes int64) bool {
+	if s.bandwidthLimiter == nil {
+		return true // No rate limiting configured
+	}
+
+	// Check if user is exempt
+	if s.isRateLimitExempt(pubkey, remoteAddr) {
+		return true
+	}
+
+	// Use pubkey hex if available, otherwise IP
+	var identity string
+	if len(pubkey) > 0 {
+		identity = string(pubkey) // Will be converted to hex in handler
+	} else {
+		identity = remoteAddr
+	}
+
+	return s.bandwidthLimiter.CheckAndConsume(identity, sizeBytes)
+}
+
 // BaseURLKey is the context key for the base URL (exported for use by app handler)
 type BaseURLKey struct{}

--- a/pkg/database/curating-acl.go
+++ b/pkg/database/curating-acl.go
@@ -965,14 +965,17 @@ func kindInRange(kind int, rangeStr string) bool {
 // kindInCategory checks if a kind belongs to a predefined category
 func kindInCategory(kind int, category string) bool {
 	categories := map[string][]int{
-		"social":      {0, 1, 3, 6, 7, 10002},
-		"dm":          {4, 14, 1059},
-		"longform":    {30023, 30024},
-		"media":       {1063, 20, 21, 22},
-		"marketplace": {30017, 30018, 30019, 30020, 1021, 1022},
-		"groups_nip29": {9, 10, 11, 12, 9000, 9001, 9002, 39000, 39001, 39002},
-		"groups_nip72": {34550, 1111, 4550},
-		"lists":       {10000, 10001, 10003, 30000, 30001, 30003},
+		"social":              {0, 1, 3, 6, 7, 10002},
+		"dm":                  {4, 14, 1059},
+		"longform":            {30023, 30024},
+		"media":               {1063, 20, 21, 22},
+		"marketplace":         {30017, 30018, 30019, 30020, 1021, 1022}, // Legacy alias
+		"marketplace_nip15":   {30017, 30018, 30019, 30020, 1021, 1022},
+		"marketplace_nip99":   {30402, 30403, 30405, 30406, 31555}, // NIP-99/Gamma Markets (Plebeian Market)
+		"order_communication": {16, 17},                            // Gamma Markets order messages
+		"groups_nip29":        {9, 10, 11, 12, 9000, 9001, 9002, 39000, 39001, 39002},
+		"groups_nip72":        {34550, 1111, 4550},
+		"lists":               {10000, 10001, 10003, 30000, 30001, 30003},
 	}

 	kinds, ok := categories[category]
--- a/pkg/database/get-indexes-from-filter.go
+++ b/pkg/database/get-indexes-from-filter.go
@@ -6,6 +6,7 @@ import (
 	"sort"

 	"lol.mleku.dev/chk"
+	"lol.mleku.dev/errorf"
 	"lol.mleku.dev/log"
 	"next.orly.dev/pkg/database/indexes"
 	types2 "next.orly.dev/pkg/database/indexes/types"
@@ -44,6 +45,12 @@ func NormalizeTagValueForHash(key byte, valueBytes []byte) []byte {
 func CreateIdHashFromData(data []byte) (i *types2.IdHash, err error) {
 	i = new(types2.IdHash)

+	// Skip empty data to avoid noisy errors
+	if len(data) == 0 {
+		err = errorf.E("CreateIdHashFromData: empty ID provided")
+		return
+	}
+
 	// If data looks like hex string and has the right length for hex-encoded
 	// sha256
 	if len(data) == 64 {
@@ -95,6 +102,11 @@ func GetIndexesFromFilter(f *filter.F) (idxs []Range, err error) {
 	// should be an error, but convention just ignores it.
 	if f.Ids.Len() > 0 {
 		for _, id := range f.Ids.T {
+			// Skip empty IDs - some filters have empty ID values
+			if len(id) == 0 {
+				log.D.F("GetIndexesFromFilter: skipping empty ID in filter (ids=%d)", f.Ids.Len())
+				continue
+			}
 			if err = func() (err error) {
 				var i *types2.IdHash
 				if i, err = CreateIdHashFromData(id); chk.E(err) {
--- a/pkg/database/get-serial-by-id.go
+++ b/pkg/database/get-serial-by-id.go
@@ -20,6 +20,10 @@ import (

 func (d *D) GetSerialById(id []byte) (ser *types.Uint40, err error) {
 	// log.T.F("GetSerialById: input id=%s", hex.Enc(id))
+	if len(id) == 0 {
+		err = errorf.E("GetSerialById: called with empty ID")
+		return
+	}
 	var idxs []Range
 	if idxs, err = GetIndexesFromFilter(&filter.F{Ids: tag.NewFromBytesSlice(id)}); chk.E(err) {
 		return
@@ -102,6 +106,10 @@ func (d *D) GetSerialsByIdsWithFilter(

 			// Process each ID sequentially
 			for _, id := range ids.T {
+				// Skip empty IDs
+				if len(id) == 0 {
+					continue
+				}
 				// idHex := hex.Enc(id)

 				// Get the index prefix for this ID
--- a/pkg/database/indexes/types/idhash.go
+++ b/pkg/database/indexes/types/idhash.go
@@ -24,8 +24,8 @@ func (i *IdHash) Set(idh []byte) {
 func (i *IdHash) FromId(id []byte) (err error) {
 	if len(id) != sha256.Size {
 		err = errorf.E(
-			"FromId: invalid ID length, got %d require %d", len(id),
-			sha256.Size,
+			"FromId: invalid ID length, got %d require %d (data=%x)", len(id),
+			sha256.Size, id,
 		)
 		return
 	}
--- a/pkg/event/routing/ephemeral.go
+++ b/pkg/event/routing/ephemeral.go
@@ -3,6 +3,7 @@ package routing
 import (
 	"git.mleku.dev/mleku/nostr/encoders/event"
 	"git.mleku.dev/mleku/nostr/encoders/kind"
+	"lol.mleku.dev/log"
 )

 // Publisher abstracts event delivery to subscribers.
@@ -22,6 +23,7 @@ func IsEphemeral(k uint16) bool {
 // - Are immediately delivered to subscribers
 func MakeEphemeralHandler(publisher Publisher) Handler {
 	return func(ev *event.E, authedPubkey []byte) Result {
+		log.I.F("ephemeral handler received event kind %d, id %0x", ev.Kind, ev.ID[:8])
 		// Clone and deliver immediately without persistence
 		cloned := ev.Clone()
 		go publisher.Deliver(cloned)
--- a/pkg/neo4j/query-events.go
+++ b/pkg/neo4j/query-events.go
@@ -10,12 +10,15 @@ import (
 	"git.mleku.dev/mleku/nostr/encoders/filter"
 	"git.mleku.dev/mleku/nostr/encoders/hex"
 	"git.mleku.dev/mleku/nostr/encoders/tag"
+	"lol.mleku.dev/log"
 	"next.orly.dev/pkg/database/indexes/types"
 	"next.orly.dev/pkg/interfaces/store"
 )

 // QueryEvents retrieves events matching the given filter
 func (n *N) QueryEvents(c context.Context, f *filter.F) (evs event.S, err error) {
+	log.T.F("Neo4j QueryEvents called with filter: kinds=%v, authors=%d, tags=%v",
+		f.Kinds != nil, f.Authors != nil && len(f.Authors.T) > 0, f.Tags != nil)
 	return n.QueryEventsWithOptions(c, f, false, false)
 }

@@ -101,6 +104,7 @@ func (n *N) buildCypherQuery(f *filter.F, includeDeleteEvents bool) (string, map
 			// Normalize to lowercase hex using our utility function
 			// This handles both binary-encoded pubkeys and hex string pubkeys (including uppercase)
 			hexAuthor := NormalizePubkeyHex(author)
+			log.T.F("Neo4j author filter: raw_len=%d, normalized=%q", len(author), hexAuthor)
 			if hexAuthor == "" {
 				continue
 			}
@@ -130,30 +134,39 @@ func (n *N) buildCypherQuery(f *filter.F, includeDeleteEvents bool) (string, map
 	}

 	// Time range filters - for temporal queries
-	if f.Since != nil {
+	// Note: Check both pointer and value - a zero timestamp (Unix epoch 1970) is almost
+	// certainly not a valid constraint as Nostr events didn't exist then
+	if f.Since != nil && f.Since.V > 0 {
 		params["since"] = f.Since.V
 		whereClauses = append(whereClauses, "e.created_at >= $since")
 	}
-	if f.Until != nil {
+	if f.Until != nil && f.Until.V > 0 {
 		params["until"] = f.Until.V
 		whereClauses = append(whereClauses, "e.created_at <= $until")
 	}

 	// Tag filters - this is where Neo4j's graph capabilities shine
-	// We can efficiently traverse tag relationships
+	// We use EXISTS subqueries to efficiently filter events by tags
+	// This ensures events are only returned if they have matching tags
 	tagIndex := 0
 	if f.Tags != nil {
 		for _, tagValues := range *f.Tags {
 			if len(tagValues.T) > 0 {
-				tagVarName := fmt.Sprintf("t%d", tagIndex)
 				tagTypeParam := fmt.Sprintf("tagType_%d", tagIndex)
 				tagValuesParam := fmt.Sprintf("tagValues_%d", tagIndex)

-				// Add tag relationship to MATCH clause
-				matchClause += fmt.Sprintf(" OPTIONAL MATCH (e)-[:TAGGED_WITH]->(%s:Tag)", tagVarName)
+				// The first element is the tag type (e.g., "e", "p", "#e", "#p", etc.)
+				// Filter tags may have "#" prefix (e.g., "#d" for d-tag filters)
+				// Event tags are stored without prefix, so we must strip it
+				tagTypeBytes := tagValues.T[0]
+				var tagType string
+				if len(tagTypeBytes) > 0 && tagTypeBytes[0] == '#' {
+					tagType = string(tagTypeBytes[1:]) // Strip "#" prefix
+				} else {
+					tagType = string(tagTypeBytes)
+				}

-				// The first element is the tag type (e.g., "e", "p", etc.)
-				tagType := string(tagValues.T[0])
+				log.T.F("Neo4j tag filter: type=%q (raw=%q, len=%d)", tagType, string(tagTypeBytes), len(tagTypeBytes))

 				// Convert remaining tag values to strings (skip first element which is the type)
 				// For e/p tags, use NormalizePubkeyHex to handle binary encoding and uppercase hex
@@ -162,26 +175,34 @@ func (n *N) buildCypherQuery(f *filter.F, includeDeleteEvents bool) (string, map
 					if tagType == "e" || tagType == "p" {
 						// Normalize e/p tag values to lowercase hex (handles binary encoding)
 						normalized := NormalizePubkeyHex(tv)
+						log.T.F("Neo4j tag filter: %s-tag value normalized: %q (raw len=%d, binary=%v)",
+							tagType, normalized, len(tv), IsBinaryEncoded(tv))
 						if normalized != "" {
 							tagValueStrings = append(tagValueStrings, normalized)
 						}
 					} else {
 						// For other tags, use direct string conversion
-						tagValueStrings = append(tagValueStrings, string(tv))
+						val := string(tv)
+						log.T.F("Neo4j tag filter: %s-tag value: %q (len=%d)", tagType, val, len(val))
+						tagValueStrings = append(tagValueStrings, val)
 					}
 				}

 				// Skip if no valid values after normalization
 				if len(tagValueStrings) == 0 {
+					log.W.F("Neo4j tag filter: no valid values for tag type %q, skipping", tagType)
 					continue
 				}

-				// Add WHERE conditions for this tag
+				log.T.F("Neo4j tag filter: type=%s, values=%v", tagType, tagValueStrings)
+
+				// Use EXISTS subquery to filter events that have matching tags
+				// This is more correct than OPTIONAL MATCH because it requires the tag to exist
 				params[tagTypeParam] = tagType
 				params[tagValuesParam] = tagValueStrings
 				whereClauses = append(whereClauses,
-					fmt.Sprintf("(%s.type = $%s AND %s.value IN $%s)",
-						tagVarName, tagTypeParam, tagVarName, tagValuesParam))
+					fmt.Sprintf("EXISTS { MATCH (e)-[:TAGGED_WITH]->(t:Tag) WHERE t.type = $%s AND t.value IN $%s }",
+						tagTypeParam, tagValuesParam))

 				tagIndex++
 			}
@@ -248,6 +269,26 @@ RETURN e.id AS id,
 	// Combine all parts
 	cypher := matchClause + whereClause + returnClause + orderClause + limitClause

+	// Log the generated query for debugging
+	log.T.F("Neo4j query: %s", cypher)
+	// Log params at trace level for debugging
+	var paramSummary strings.Builder
+	for k, v := range params {
+		switch val := v.(type) {
+		case []string:
+			if len(val) <= 3 {
+				paramSummary.WriteString(fmt.Sprintf("%s: %v ", k, val))
+			} else {
+				paramSummary.WriteString(fmt.Sprintf("%s: [%d values] ", k, len(val)))
+			}
+		case []int64:
+			paramSummary.WriteString(fmt.Sprintf("%s: %v ", k, val))
+		default:
+			paramSummary.WriteString(fmt.Sprintf("%s: %v ", k, v))
+		}
+	}
+	log.T.F("Neo4j params: %s", paramSummary.String())
+
 	return cypher, params
 }

@@ -300,19 +341,17 @@ func (n *N) parseEventsFromResult(result *CollectedResult) ([]*event.E, error) {
 			_ = tags.UnmarshalJSON([]byte(tagsStr))
 		}

-		// Create event
+		// Create event with decoded binary fields
 		e := &event.E{
+			ID:        id,
+			Pubkey:    pubkey,
 			Kind:      uint16(kind),
 			CreatedAt: createdAt,
 			Content:   []byte(content),
 			Tags:      tags,
+			Sig:       sig,
 		}

-		// Copy fixed-size arrays
-		copy(e.ID[:], id)
-		copy(e.Sig[:], sig)
-		copy(e.Pubkey[:], pubkey)
-
 		events = append(events, e)
 	}

--- a/pkg/neo4j/query-events_test.go
+++ b/pkg/neo4j/query-events_test.go
@@ -462,3 +462,584 @@ func TestCountEvents(t *testing.T) {

 	t.Logf("✓ Count events returned correct count: %d", count)
 }
+
+// TestQueryEventsByTagWithHashPrefix tests that tag filters with "#" prefix work correctly.
+// This is a regression test for a bug where filter tags like "#d" were not being matched
+// because the "#" prefix wasn't being stripped before comparison with stored tags.
+func TestQueryEventsByTagWithHashPrefix(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create events with d-tags (parameterized replaceable kind)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with d=id1",
+		tag.NewS(tag.NewFromAny("d", "id1")), baseTs)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with d=id2",
+		tag.NewS(tag.NewFromAny("d", "id2")), baseTs+1)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with d=id3",
+		tag.NewS(tag.NewFromAny("d", "id3")), baseTs+2)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with d=other",
+		tag.NewS(tag.NewFromAny("d", "other")), baseTs+3)
+
+	// Query with "#d" prefix (as clients send it) - should match events with d=id1
+	evs, err := testDB.QueryEvents(ctx, &filter.F{
+		Kinds: kind.NewS(kind.New(30382)),
+		Tags:  tag.NewS(tag.NewFromAny("#d", "id1")),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query events with #d tag: %v", err)
+	}
+
+	if len(evs) != 1 {
+		t.Fatalf("Expected 1 event with d=id1, got %d", len(evs))
+	}
+
+	// Verify the returned event has the correct d-tag
+	dTag := evs[0].Tags.GetFirst([]byte("d"))
+	if dTag == nil || string(dTag.Value()) != "id1" {
+		t.Fatalf("Expected d=id1, got d=%s", dTag.Value())
+	}
+
+	t.Logf("✓ Query with #d prefix returned correct event")
+}
+
+// TestQueryEventsByTagMultipleValues tests that tag filters with multiple values
+// use OR logic (match events with ANY of the values).
+func TestQueryEventsByTagMultipleValues(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create events with different d-tags
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event A",
+		tag.NewS(tag.NewFromAny("d", "target-1")), baseTs)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event B",
+		tag.NewS(tag.NewFromAny("d", "target-2")), baseTs+1)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event C",
+		tag.NewS(tag.NewFromAny("d", "target-3")), baseTs+2)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event D (not target)",
+		tag.NewS(tag.NewFromAny("d", "other-value")), baseTs+3)
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event E (no match)",
+		tag.NewS(tag.NewFromAny("d", "different")), baseTs+4)
+
+	// Query with multiple d-tag values using "#d" prefix
+	// Should match events with d=target-1 OR d=target-2 OR d=target-3
+	evs, err := testDB.QueryEvents(ctx, &filter.F{
+		Kinds: kind.NewS(kind.New(30382)),
+		Tags:  tag.NewS(tag.NewFromAny("#d", "target-1", "target-2", "target-3")),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query events with multiple #d values: %v", err)
+	}
+
+	if len(evs) != 3 {
+		t.Fatalf("Expected 3 events matching the d-tag values, got %d", len(evs))
+	}
+
+	// Verify returned events have correct d-tags
+	validDTags := map[string]bool{"target-1": false, "target-2": false, "target-3": false}
+	for _, ev := range evs {
+		dTag := ev.Tags.GetFirst([]byte("d"))
+		if dTag == nil {
+			t.Fatalf("Event missing d-tag")
+		}
+		dValue := string(dTag.Value())
+		if _, ok := validDTags[dValue]; !ok {
+			t.Fatalf("Unexpected d-tag value: %s", dValue)
+		}
+		validDTags[dValue] = true
+	}
+
+	// Verify all expected d-tags were found
+	for dValue, found := range validDTags {
+		if !found {
+			t.Fatalf("Expected to find event with d=%s", dValue)
+		}
+	}
+
+	t.Logf("✓ Query with multiple #d values returned correct events")
+}
+
+// TestQueryEventsByTagNoMatch tests that tag filters correctly return no results
+// when no events match the filter.
+func TestQueryEventsByTagNoMatch(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create events with d-tags
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event",
+		tag.NewS(tag.NewFromAny("d", "existing-value")), baseTs)
+
+	// Query for d-tag value that doesn't exist
+	evs, err := testDB.QueryEvents(ctx, &filter.F{
+		Kinds: kind.NewS(kind.New(30382)),
+		Tags:  tag.NewS(tag.NewFromAny("#d", "non-existent-value")),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query events: %v", err)
+	}
+
+	if len(evs) != 0 {
+		t.Fatalf("Expected 0 events for non-matching d-tag, got %d", len(evs))
+	}
+
+	t.Logf("✓ Query with non-matching #d value returned no events")
+}
+
+// TestQueryEventsByTagWithKindAndAuthor tests the combination of kind, author, and tag filters.
+// This is the specific case reported by the user with kind 30382.
+func TestQueryEventsByTagWithKindAndAuthor(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	alice := createTestSignerLocal(t)
+	bob := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create events from different authors with d-tags
+	createAndSaveEventLocal(t, ctx, alice, 30382, "Alice target 1",
+		tag.NewS(tag.NewFromAny("d", "card-1")), baseTs)
+	createAndSaveEventLocal(t, ctx, alice, 30382, "Alice target 2",
+		tag.NewS(tag.NewFromAny("d", "card-2")), baseTs+1)
+	createAndSaveEventLocal(t, ctx, alice, 30382, "Alice other",
+		tag.NewS(tag.NewFromAny("d", "other-card")), baseTs+2)
+	createAndSaveEventLocal(t, ctx, bob, 30382, "Bob target 1",
+		tag.NewS(tag.NewFromAny("d", "card-1")), baseTs+3) // Same d-tag as Alice but different author
+
+	// Query for Alice's events with specific d-tags
+	evs, err := testDB.QueryEvents(ctx, &filter.F{
+		Kinds:   kind.NewS(kind.New(30382)),
+		Authors: tag.NewFromBytesSlice(alice.Pub()),
+		Tags:    tag.NewS(tag.NewFromAny("#d", "card-1", "card-2")),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query events: %v", err)
+	}
+
+	// Should only return Alice's 2 events, not Bob's even though he has card-1
+	if len(evs) != 2 {
+		t.Fatalf("Expected 2 events from Alice with matching d-tags, got %d", len(evs))
+	}
+
+	alicePubkey := hex.Enc(alice.Pub())
+	for _, ev := range evs {
+		if hex.Enc(ev.Pubkey[:]) != alicePubkey {
+			t.Fatalf("Expected author %s, got %s", alicePubkey, hex.Enc(ev.Pubkey[:]))
+		}
+		dTag := ev.Tags.GetFirst([]byte("d"))
+		dValue := string(dTag.Value())
+		if dValue != "card-1" && dValue != "card-2" {
+			t.Fatalf("Expected d=card-1 or card-2, got d=%s", dValue)
+		}
+	}
+
+	t.Logf("✓ Query with kind, author, and #d filter returned correct events")
+}
+
+// TestBinaryTagFilterRegression tests that queries with #e and #p tags work correctly
+// even when tags are stored with binary-encoded values but filters come as hex strings.
+// This mirrors the Badger database test for binary tag handling.
+func TestBinaryTagFilterRegression(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	author := createTestSignerLocal(t)
+	referenced := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create a referenced event to get a valid event ID for e-tag
+	refEvent := createAndSaveEventLocal(t, ctx, referenced, 1, "Referenced event", nil, baseTs)
+
+	// Get hex representations
+	refEventIdHex := hex.Enc(refEvent.ID)
+	refPubkeyHex := hex.Enc(referenced.Pub())
+
+	// Create test event with e, p, d, and other tags
+	testEvent := createAndSaveEventLocal(t, ctx, author, 30520, "Event with binary tags",
+		tag.NewS(
+			tag.NewFromAny("d", "test-d-value"),
+			tag.NewFromAny("p", string(refPubkeyHex)),
+			tag.NewFromAny("e", string(refEventIdHex)),
+			tag.NewFromAny("t", "test-topic"),
+		), baseTs+1)
+
+	testEventIdHex := hex.Enc(testEvent.ID)
+
+	// Test case 1: Query WITHOUT #e/#p tags (baseline - should work)
+	t.Run("QueryWithoutEPTags", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds:   kind.NewS(kind.New(30520)),
+			Authors: tag.NewFromBytesSlice(author.Pub()),
+			Tags:    tag.NewS(tag.NewFromAny("#d", "test-d-value")),
+		})
+		if err != nil {
+			t.Fatalf("Query without e/p tags failed: %v", err)
+		}
+
+		if len(evs) == 0 {
+			t.Fatal("Expected to find event with d tag filter, got 0 results")
+		}
+
+		found := false
+		for _, ev := range evs {
+			if hex.Enc(ev.ID) == testEventIdHex {
+				found = true
+				break
+			}
+		}
+		if !found {
+			t.Errorf("Expected event ID %s not found", testEventIdHex)
+		}
+	})
+
+	// Test case 2: Query WITH #p tag
+	t.Run("QueryWithPTag", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds:   kind.NewS(kind.New(30520)),
+			Authors: tag.NewFromBytesSlice(author.Pub()),
+			Tags: tag.NewS(
+				tag.NewFromAny("#d", "test-d-value"),
+				tag.NewFromAny("#p", string(refPubkeyHex)),
+			),
+		})
+		if err != nil {
+			t.Fatalf("Query with #p tag failed: %v", err)
+		}
+
+		if len(evs) == 0 {
+			t.Fatalf("REGRESSION: Expected to find event with #p tag filter, got 0 results")
+		}
+	})
+
+	// Test case 3: Query WITH #e tag
+	t.Run("QueryWithETag", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds:   kind.NewS(kind.New(30520)),
+			Authors: tag.NewFromBytesSlice(author.Pub()),
+			Tags: tag.NewS(
+				tag.NewFromAny("#d", "test-d-value"),
+				tag.NewFromAny("#e", string(refEventIdHex)),
+			),
+		})
+		if err != nil {
+			t.Fatalf("Query with #e tag failed: %v", err)
+		}
+
+		if len(evs) == 0 {
+			t.Fatalf("REGRESSION: Expected to find event with #e tag filter, got 0 results")
+		}
+	})
+
+	// Test case 4: Query WITH BOTH #e AND #p tags
+	t.Run("QueryWithBothEAndPTags", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds:   kind.NewS(kind.New(30520)),
+			Authors: tag.NewFromBytesSlice(author.Pub()),
+			Tags: tag.NewS(
+				tag.NewFromAny("#d", "test-d-value"),
+				tag.NewFromAny("#e", string(refEventIdHex)),
+				tag.NewFromAny("#p", string(refPubkeyHex)),
+			),
+		})
+		if err != nil {
+			t.Fatalf("Query with both #e and #p tags failed: %v", err)
+		}
+
+		if len(evs) == 0 {
+			t.Fatalf("REGRESSION: Expected to find event with #e and #p tag filters, got 0 results")
+		}
+	})
+
+	t.Logf("✓ Binary tag filter regression tests passed")
+}
+
+// TestParameterizedReplaceableEvents tests that parameterized replaceable events (kind 30000+)
+// are handled correctly - only the newest version should be returned in queries by kind/author/d-tag.
+func TestParameterizedReplaceableEvents(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create older parameterized replaceable event
+	createAndSaveEventLocal(t, ctx, signer, 30000, "Original event",
+		tag.NewS(tag.NewFromAny("d", "test-param")), baseTs-7200) // 2 hours ago
+
+	// Create newer event with same kind/author/d-tag
+	createAndSaveEventLocal(t, ctx, signer, 30000, "Newer event",
+		tag.NewS(tag.NewFromAny("d", "test-param")), baseTs-3600) // 1 hour ago
+
+	// Create newest event with same kind/author/d-tag
+	newestEvent := createAndSaveEventLocal(t, ctx, signer, 30000, "Newest event",
+		tag.NewS(tag.NewFromAny("d", "test-param")), baseTs) // Now
+
+	// Query for events - should only return the newest one
+	evs, err := testDB.QueryEvents(ctx, &filter.F{
+		Kinds:   kind.NewS(kind.New(30000)),
+		Authors: tag.NewFromBytesSlice(signer.Pub()),
+		Tags:    tag.NewS(tag.NewFromAny("#d", "test-param")),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query parameterized replaceable events: %v", err)
+	}
+
+	// Note: Neo4j backend may or may not automatically deduplicate replaceable events
+	// depending on implementation. The important thing is that the newest is returned first.
+	if len(evs) == 0 {
+		t.Fatal("Expected at least 1 event")
+	}
+
+	// Verify the first (most recent) event is the newest one
+	if hex.Enc(evs[0].ID) != hex.Enc(newestEvent.ID) {
+		t.Logf("Note: Expected newest event first, got different order")
+	}
+
+	t.Logf("✓ Parameterized replaceable events test returned %d events", len(evs))
+}
+
+// TestQueryForIds tests the QueryForIds method
+func TestQueryForIds(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create test events
+	ev1 := createAndSaveEventLocal(t, ctx, signer, 1, "Event 1", nil, baseTs)
+	ev2 := createAndSaveEventLocal(t, ctx, signer, 1, "Event 2", nil, baseTs+1)
+	createAndSaveEventLocal(t, ctx, signer, 7, "Reaction", nil, baseTs+2)
+
+	// Query for IDs of kind 1 events
+	idPkTs, err := testDB.QueryForIds(ctx, &filter.F{
+		Kinds: kind.NewS(kind.New(1)),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query for IDs: %v", err)
+	}
+
+	if len(idPkTs) != 2 {
+		t.Fatalf("Expected 2 IDs for kind 1 events, got %d", len(idPkTs))
+	}
+
+	// Verify IDs match our events
+	foundIds := make(map[string]bool)
+	for _, r := range idPkTs {
+		foundIds[hex.Enc(r.Id)] = true
+	}
+
+	if !foundIds[hex.Enc(ev1.ID)] {
+		t.Error("Event 1 ID not found in results")
+	}
+	if !foundIds[hex.Enc(ev2.ID)] {
+		t.Error("Event 2 ID not found in results")
+	}
+
+	t.Logf("✓ QueryForIds returned correct IDs")
+}
+
+// TestQueryForSerials tests the QueryForSerials method
+func TestQueryForSerials(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create test events
+	createAndSaveEventLocal(t, ctx, signer, 1, "Event 1", nil, baseTs)
+	createAndSaveEventLocal(t, ctx, signer, 1, "Event 2", nil, baseTs+1)
+	createAndSaveEventLocal(t, ctx, signer, 1, "Event 3", nil, baseTs+2)
+
+	// Query for serials
+	serials, err := testDB.QueryForSerials(ctx, &filter.F{
+		Kinds: kind.NewS(kind.New(1)),
+	})
+	if err != nil {
+		t.Fatalf("Failed to query for serials: %v", err)
+	}
+
+	if len(serials) != 3 {
+		t.Fatalf("Expected 3 serials, got %d", len(serials))
+	}
+
+	t.Logf("✓ QueryForSerials returned %d serials", len(serials))
+}
+
+// TestQueryEventsComplex tests complex filter combinations
+func TestQueryEventsComplex(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	alice := createTestSignerLocal(t)
+	bob := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create diverse set of events
+	createAndSaveEventLocal(t, ctx, alice, 1, "Alice note with bitcoin tag",
+		tag.NewS(tag.NewFromAny("t", "bitcoin")), baseTs)
+	createAndSaveEventLocal(t, ctx, alice, 1, "Alice note with nostr tag",
+		tag.NewS(tag.NewFromAny("t", "nostr")), baseTs+1)
+	createAndSaveEventLocal(t, ctx, alice, 7, "Alice reaction",
+		nil, baseTs+2)
+	createAndSaveEventLocal(t, ctx, bob, 1, "Bob note with bitcoin tag",
+		tag.NewS(tag.NewFromAny("t", "bitcoin")), baseTs+3)
+
+	// Test: kinds + tags (no authors)
+	t.Run("KindsAndTags", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds: kind.NewS(kind.New(1)),
+			Tags:  tag.NewS(tag.NewFromAny("#t", "bitcoin")),
+		})
+		if err != nil {
+			t.Fatalf("Query failed: %v", err)
+		}
+		if len(evs) != 2 {
+			t.Fatalf("Expected 2 events with kind=1 and #t=bitcoin, got %d", len(evs))
+		}
+	})
+
+	// Test: authors + tags (no kinds)
+	t.Run("AuthorsAndTags", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Authors: tag.NewFromBytesSlice(alice.Pub()),
+			Tags:    tag.NewS(tag.NewFromAny("#t", "bitcoin")),
+		})
+		if err != nil {
+			t.Fatalf("Query failed: %v", err)
+		}
+		if len(evs) != 1 {
+			t.Fatalf("Expected 1 event from Alice with #t=bitcoin, got %d", len(evs))
+		}
+	})
+
+	// Test: kinds + authors (no tags)
+	t.Run("KindsAndAuthors", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds:   kind.NewS(kind.New(1)),
+			Authors: tag.NewFromBytesSlice(alice.Pub()),
+		})
+		if err != nil {
+			t.Fatalf("Query failed: %v", err)
+		}
+		if len(evs) != 2 {
+			t.Fatalf("Expected 2 kind=1 events from Alice, got %d", len(evs))
+		}
+	})
+
+	// Test: all three filters
+	t.Run("AllFilters", func(t *testing.T) {
+		evs, err := testDB.QueryEvents(ctx, &filter.F{
+			Kinds:   kind.NewS(kind.New(1)),
+			Authors: tag.NewFromBytesSlice(alice.Pub()),
+			Tags:    tag.NewS(tag.NewFromAny("#t", "nostr")),
+		})
+		if err != nil {
+			t.Fatalf("Query failed: %v", err)
+		}
+		if len(evs) != 1 {
+			t.Fatalf("Expected 1 event (Alice kind=1 #t=nostr), got %d", len(evs))
+		}
+	})
+
+	t.Logf("✓ Complex filter combination tests passed")
+}
+
+// TestQueryEventsMultipleTagTypes tests filtering with multiple different tag types
+func TestQueryEventsMultipleTagTypes(t *testing.T) {
+	if testDB == nil {
+		t.Skip("Neo4j not available")
+	}
+
+	cleanTestDatabase()
+
+	ctx := context.Background()
+	signer := createTestSignerLocal(t)
+	baseTs := timestamp.Now().V
+
+	// Create events with multiple tag types
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with d and client tags",
+		tag.NewS(
+			tag.NewFromAny("d", "user-1"),
+			tag.NewFromAny("client", "app-a"),
+		), baseTs)
+
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with d and different client",
+		tag.NewS(
+			tag.NewFromAny("d", "user-2"),
+			tag.NewFromAny("client", "app-b"),
+		), baseTs+1)
+
+	createAndSaveEventLocal(t, ctx, signer, 30382, "Event with only d tag",
+		tag.NewS(
+			tag.NewFromAny("d", "user-3"),
+		), baseTs+2)
+
+	// Query with multiple tag types (should AND them together)
+	evs, err := testDB.QueryEvents(ctx, &filter.F{
+		Kinds: kind.NewS(kind.New(30382)),
+		Tags: tag.NewS(
+			tag.NewFromAny("#d", "user-1", "user-2"),
+			tag.NewFromAny("#client", "app-a"),
+		),
+	})
+	if err != nil {
+		t.Fatalf("Query with multiple tag types failed: %v", err)
+	}
+
+	// Should match only the first event (user-1 with app-a)
+	if len(evs) != 1 {
+		t.Fatalf("Expected 1 event matching both #d and #client, got %d", len(evs))
+	}
+
+	dTag := evs[0].Tags.GetFirst([]byte("d"))
+	if string(dTag.Value()) != "user-1" {
+		t.Fatalf("Expected d=user-1, got d=%s", dTag.Value())
+	}
+
+	t.Logf("✓ Multiple tag types filter test passed")
+}
--- a/pkg/ratelimit/factory.go
+++ b/pkg/ratelimit/factory.go
@@ -54,3 +54,11 @@ func MonitorFromNeo4jDriver(
 ) loadmonitor.Monitor {
 	return NewNeo4jMonitor(driver, querySem, maxConcurrency, 100*time.Millisecond)
 }
+
+// NewMemoryOnlyLimiter creates a rate limiter that only monitors process memory.
+// Useful for database backends that don't have their own load metrics (e.g., BBolt).
+// Since BBolt uses memory-mapped IO, memory pressure is still relevant.
+func NewMemoryOnlyLimiter(config Config) *Limiter {
+	monitor := NewMemoryMonitor(100 * time.Millisecond)
+	return NewLimiter(config, monitor)
+}
--- a/pkg/ratelimit/limiter.go
+++ b/pkg/ratelimit/limiter.go
@@ -377,29 +377,26 @@ func (l *Limiter) ComputeDelay(opType OperationType) time.Duration {

 		// In emergency mode, apply progressive throttling for writes
 		if inEmergency {
-			// Calculate how far above recovery threshold we are
-			// At emergency threshold, add 1x normal delay
-			// For every additional 10% above emergency, double the delay
-			excessPressure := metrics.MemoryPressure - l.config.RecoveryThreshold
-			if excessPressure > 0 {
-				// Progressive multiplier: starts at 2x, doubles every 10% excess
-				multiplier := 2.0
-				for excess := excessPressure; excess > 0.1; excess -= 0.1 {
-					multiplier *= 2
-				}
-
-				emergencyDelaySec := delaySec * multiplier
-				maxEmergencySec := float64(l.config.EmergencyMaxDelayMs) / 1000.0
-
-				if emergencyDelaySec > maxEmergencySec {
-					emergencyDelaySec = maxEmergencySec
-				}
-				// Minimum emergency delay of 100ms to allow other operations
-				if emergencyDelaySec < 0.1 {
-					emergencyDelaySec = 0.1
-				}
-				delaySec = emergencyDelaySec
+			// Calculate how far above emergency threshold we are
+			// Linear scaling: multiplier = 1 + (excess * 5)
+			// At emergency threshold: 1x, at +20% above: 2x, at +40% above: 3x
+			excessPressure := metrics.MemoryPressure - l.config.EmergencyThreshold
+			if excessPressure < 0 {
+				excessPressure = 0
 			}
+			multiplier := 1.0 + excessPressure*5.0
+
+			emergencyDelaySec := delaySec * multiplier
+			maxEmergencySec := float64(l.config.EmergencyMaxDelayMs) / 1000.0
+
+			if emergencyDelaySec > maxEmergencySec {
+				emergencyDelaySec = maxEmergencySec
+			}
+			// Minimum emergency delay of 100ms to allow other operations
+			if emergencyDelaySec < 0.1 {
+				emergencyDelaySec = 0.1
+			}
+			delaySec = emergencyDelaySec
 		}

 		if delaySec > 0 {
--- a/pkg/ratelimit/memory_monitor.go
+++ b/pkg/ratelimit/memory_monitor.go
@@ -0,0 +1,214 @@
+//go:build !(js && wasm)
+
+package ratelimit
+
+import (
+	"sync"
+	"sync/atomic"
+	"time"
+
+	"next.orly.dev/pkg/interfaces/loadmonitor"
+)
+
+// MemoryMonitor is a simple load monitor that only tracks process memory.
+// Used for database backends that don't have their own load metrics (e.g., BBolt).
+type MemoryMonitor struct {
+	// Configuration
+	pollInterval time.Duration
+	targetBytes  atomic.Uint64
+
+	// State
+	running  atomic.Bool
+	stopChan chan struct{}
+	doneChan chan struct{}
+
+	// Metrics (protected by mutex)
+	mu             sync.RWMutex
+	currentMetrics loadmonitor.Metrics
+
+	// Latency tracking
+	queryLatencies []time.Duration
+	writeLatencies []time.Duration
+	latencyMu      sync.Mutex
+
+	// Emergency mode
+	emergencyThreshold float64 // e.g., 1.167 (target + 1/6)
+	recoveryThreshold  float64 // e.g., 0.833 (target - 1/6)
+	inEmergency        atomic.Bool
+}
+
+// NewMemoryMonitor creates a memory-only load monitor.
+// pollInterval controls how often memory is sampled (recommended: 100ms).
+func NewMemoryMonitor(pollInterval time.Duration) *MemoryMonitor {
+	m := &MemoryMonitor{
+		pollInterval:       pollInterval,
+		stopChan:           make(chan struct{}),
+		doneChan:           make(chan struct{}),
+		queryLatencies:     make([]time.Duration, 0, 100),
+		writeLatencies:     make([]time.Duration, 0, 100),
+		emergencyThreshold: 1.167, // Default: target + 1/6
+		recoveryThreshold:  0.833, // Default: target - 1/6
+	}
+	return m
+}
+
+// GetMetrics returns the current load metrics.
+func (m *MemoryMonitor) GetMetrics() loadmonitor.Metrics {
+	m.mu.RLock()
+	defer m.mu.RUnlock()
+	return m.currentMetrics
+}
+
+// RecordQueryLatency records a query latency sample.
+func (m *MemoryMonitor) RecordQueryLatency(latency time.Duration) {
+	m.latencyMu.Lock()
+	defer m.latencyMu.Unlock()
+
+	m.queryLatencies = append(m.queryLatencies, latency)
+	if len(m.queryLatencies) > 100 {
+		m.queryLatencies = m.queryLatencies[1:]
+	}
+}
+
+// RecordWriteLatency records a write latency sample.
+func (m *MemoryMonitor) RecordWriteLatency(latency time.Duration) {
+	m.latencyMu.Lock()
+	defer m.latencyMu.Unlock()
+
+	m.writeLatencies = append(m.writeLatencies, latency)
+	if len(m.writeLatencies) > 100 {
+		m.writeLatencies = m.writeLatencies[1:]
+	}
+}
+
+// SetMemoryTarget sets the target memory limit in bytes.
+func (m *MemoryMonitor) SetMemoryTarget(bytes uint64) {
+	m.targetBytes.Store(bytes)
+}
+
+// SetEmergencyThreshold sets the memory threshold for emergency mode.
+func (m *MemoryMonitor) SetEmergencyThreshold(threshold float64) {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	m.emergencyThreshold = threshold
+}
+
+// GetEmergencyThreshold returns the current emergency threshold.
+func (m *MemoryMonitor) GetEmergencyThreshold() float64 {
+	m.mu.RLock()
+	defer m.mu.RUnlock()
+	return m.emergencyThreshold
+}
+
+// ForceEmergencyMode manually triggers emergency mode for a duration.
+func (m *MemoryMonitor) ForceEmergencyMode(duration time.Duration) {
+	m.inEmergency.Store(true)
+	go func() {
+		time.Sleep(duration)
+		m.inEmergency.Store(false)
+	}()
+}
+
+// Start begins background metric collection.
+func (m *MemoryMonitor) Start() <-chan struct{} {
+	if m.running.Swap(true) {
+		// Already running
+		return m.doneChan
+	}
+
+	go m.pollLoop()
+	return m.doneChan
+}
+
+// Stop halts background metric collection.
+func (m *MemoryMonitor) Stop() {
+	if !m.running.Swap(false) {
+		return
+	}
+	close(m.stopChan)
+	<-m.doneChan
+}
+
+// pollLoop continuously samples memory and updates metrics.
+func (m *MemoryMonitor) pollLoop() {
+	defer close(m.doneChan)
+
+	ticker := time.NewTicker(m.pollInterval)
+	defer ticker.Stop()
+
+	for {
+		select {
+		case <-m.stopChan:
+			return
+		case <-ticker.C:
+			m.updateMetrics()
+		}
+	}
+}
+
+// updateMetrics samples current memory and updates the metrics.
+func (m *MemoryMonitor) updateMetrics() {
+	target := m.targetBytes.Load()
+	if target == 0 {
+		target = 1 // Avoid division by zero
+	}
+
+	// Get physical memory using the same method as other monitors
+	procMem := ReadProcessMemoryStats()
+	physicalMemBytes := procMem.PhysicalMemoryBytes()
+	physicalMemMB := physicalMemBytes / (1024 * 1024)
+
+	// Calculate memory pressure
+	memPressure := float64(physicalMemBytes) / float64(target)
+
+	// Check emergency mode thresholds
+	m.mu.RLock()
+	emergencyThreshold := m.emergencyThreshold
+	recoveryThreshold := m.recoveryThreshold
+	m.mu.RUnlock()
+
+	wasEmergency := m.inEmergency.Load()
+	if memPressure > emergencyThreshold {
+		m.inEmergency.Store(true)
+	} else if memPressure < recoveryThreshold && wasEmergency {
+		m.inEmergency.Store(false)
+	}
+
+	// Calculate average latencies
+	m.latencyMu.Lock()
+	var avgQuery, avgWrite time.Duration
+	if len(m.queryLatencies) > 0 {
+		var total time.Duration
+		for _, l := range m.queryLatencies {
+			total += l
+		}
+		avgQuery = total / time.Duration(len(m.queryLatencies))
+	}
+	if len(m.writeLatencies) > 0 {
+		var total time.Duration
+		for _, l := range m.writeLatencies {
+			total += l
+		}
+		avgWrite = total / time.Duration(len(m.writeLatencies))
+	}
+	m.latencyMu.Unlock()
+
+	// Update metrics
+	m.mu.Lock()
+	m.currentMetrics = loadmonitor.Metrics{
+		MemoryPressure:    memPressure,
+		WriteLoad:         0,    // No database-specific load metric
+		ReadLoad:          0,    // No database-specific load metric
+		QueryLatency:      avgQuery,
+		WriteLatency:      avgWrite,
+		Timestamp:         time.Now(),
+		InEmergencyMode:   m.inEmergency.Load(),
+		CompactionPending: false, // BBolt doesn't have compaction
+		PhysicalMemoryMB:  physicalMemMB,
+	}
+	m.mu.Unlock()
+}
+
+// Ensure MemoryMonitor implements the required interfaces
+var _ loadmonitor.Monitor = (*MemoryMonitor)(nil)
+var _ loadmonitor.EmergencyModeMonitor = (*MemoryMonitor)(nil)
--- a/pkg/version/version
+++ b/pkg/version/version
@@ -1 +1 @@
-v0.48.11
+v0.49.2
Author	SHA1	Message	Date
woikos	e68916ca5d	Fix Neo4j tag filter returning all events instead of filtering (v0.49.2) Some checks are pending Go / build-and-release (push) Waiting to run Details - Change OPTIONAL MATCH to EXISTS subquery for tag filtering in Neo4j - OPTIONAL MATCH returned rows even when tags didn't match (NULL values) - EXISTS subquery correctly requires matching tags to exist - Strip "#" prefix from filter tag types before matching - Filters use "#d", "#p", "#e" but events store tags without prefix - Add trace-level logging for Neo4j query debugging - Add comprehensive tests for Neo4j query builder - Clean up temporary debug logging from handle-req.go Files modified: - pkg/neo4j/query-events.go: Fix tag filtering with EXISTS subquery - pkg/neo4j/query-events_test.go: Add query builder tests - app/handle-req.go: Remove debug logging - pkg/version/version: Bump to v0.49.2 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-14 17:49:46 +01:00
woikos	0e30f7a697	feat: add NIP-99/Gamma Markets kind categories for curating mode Add marketplace_nip99 category with Plebeian Market event kinds: - 30402 (Products) - 30403 (Orders - legacy) - 30405 (Collections) - 30406 (Shipping options) - 31555 (Product reviews) Add order_communication category for Gamma Markets (kinds 16, 17). Rename existing marketplace category to marketplace_nip15 for clarity while keeping backward compatibility with the legacy alias. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-14 17:16:57 +01:00
woikos	a0af5bb45e	Fix Neo4j query returning zero events for REQ filters (v0.49.1) Some checks are pending Go / build-and-release (push) Waiting to run Details - Fix zero-value timestamp filter bug: since/until with value 0 were being added as WHERE clauses, causing queries to match no events - Fix event parsing: use direct slice assignment instead of copy() on nil slices for ID, Pubkey, and Sig fields Files modified: - pkg/neo4j/query-events.go: Fix buildCypherQuery and parseEventsFromResult Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-14 10:56:31 +01:00
woikos	9da1784b1b	Add Blossom bandwidth limiting and tune rate limiters (v0.49.0) Some checks are pending Go / build-and-release (push) Waiting to run Details - Add token-bucket bandwidth rate limiting for Blossom uploads - ORLY_BLOSSOM_RATE_LIMIT enables limiting (default: false) - ORLY_BLOSSOM_DAILY_LIMIT_MB sets daily limit (default: 10MB) - ORLY_BLOSSOM_BURST_LIMIT_MB sets burst cap (default: 50MB) - Followed users, admins, owners are exempt (unlimited) - Change emergency mode throttling from exponential to linear scaling - Old: 16x multiplier at emergency threshold entry - New: 1x at threshold, +1x per 20% excess pressure - Reduce follows ACL throttle increment from 200ms to 25ms per event - Update dependencies Files modified: - app/blossom.go: Pass rate limit config to blossom server - app/config/config.go: Add Blossom rate limit config options - pkg/blossom/ratelimit.go: New bandwidth limiter implementation - pkg/blossom/server.go: Add rate limiter integration - pkg/blossom/handlers.go: Check rate limits on upload/mirror/media - pkg/ratelimit/limiter.go: Linear emergency throttling - pkg/acl/follows.go: Reduce default throttle increment Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-14 08:09:39 +01:00
woikos	205f23fc0c	Add message segmentation to NRC protocol spec (v0.48.15) Some checks failed Go / build-and-release (push) Has been cancelled Details - Add CHUNK response type for large payload handling - Document chunking threshold (40KB) accounting for encryption overhead - Specify chunk message format with messageId, index, total, data fields - Add sender chunking process with Base64 encoding steps - Add receiver reassembly process with buffer management - Document 60-second timeout for incomplete chunk buffers - Update client/bridge implementation notes with chunking requirements - Add Smesh as reference implementation for client-side chunking Files modified: - docs/NIP-NRC.md: Added Message Segmentation section and updated impl notes - pkg/version/version: v0.48.14 -> v0.48.15 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-11 11:29:31 +01:00
woikos	489b9f4593	Improve release command VPS deployment docs (v0.48.14) Some checks failed Go / build-and-release (push) Has been cancelled Details - Clarify ARM64 build-on-remote approach for relay.orly.dev - Remove unnecessary git stash from deployment command - Add note about setcap needing reapplication after binary rebuild - Use explicit GOPATH and go binary path for clarity Files modified: - .claude/commands/release.md: Improved deployment step documentation - pkg/version/version: v0.48.13 -> v0.48.14 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-11 11:14:20 +01:00
woikos	604d759a6a	Fix web UI not showing cached events and add Blossom toggle (v0.48.13) Some checks failed Go / build-and-release (push) Has been cancelled Details - Fix fetchEvents() discarding IndexedDB cached events instead of merging with relay results - Add mergeAndDeduplicateEvents() helper to combine and dedupe events by ID - Add ORLY_BLOSSOM_ENABLED config option to disable Blossom server - Make fetch-kinds.js fall back to existing eventKinds.js when network unavailable Files modified: - app/web/src/nostr.js: Fix event caching, add merge helper - app/web/scripts/fetch-kinds.js: Add fallback for network failures - app/config/config.go: Add BlossomEnabled config field - app/main.go: Check BlossomEnabled before initializing Blossom server - pkg/version/version: Bump to v0.48.13 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>	2026-01-11 04:55:55 +01:00
woikos	be72b694eb	Add BBolt rate limiting and tune Badger defaults for large archives (v0.48.12) Some checks failed Go / build-and-release (push) Has been cancelled Details - Increase Badger cache defaults: block 512→1024MB, index 256→512MB - Increase serial cache defaults: pubkeys 100k→250k, event IDs 500k→1M - Change ZSTD default from level 1 (fast) to level 3 (balanced) - Add memory-only rate limiter for BBolt backend - Add BBolt to database backend docs with scaling recommendations - Document migration between Badger and BBolt backends Files modified: - app/config/config.go: Tuned defaults for large-scale deployments - main.go: Add BBolt rate limiter support - pkg/ratelimit/factory.go: Add NewMemoryOnlyLimiter factory - pkg/ratelimit/memory_monitor.go: New memory-only load monitor - CLAUDE.md: Add BBolt docs and scaling guide 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-01-10 11:55:07 +01:00