feat: add memory vector search

src/config/schema.ts

@@ -115,6 +115,23 @@ const FIELD_LABELS: Record<string, string> = {
   "gateway.reload.mode": "Config Reload Mode",
   "gateway.reload.debounceMs": "Config Reload Debounce (ms)",
   "agents.defaults.workspace": "Workspace",
+  "agents.defaults.memorySearch": "Memory Search",
+  "agents.defaults.memorySearch.enabled": "Enable Memory Search",
+  "agents.defaults.memorySearch.provider": "Memory Search Provider",
+  "agents.defaults.memorySearch.model": "Memory Search Model",
+  "agents.defaults.memorySearch.fallback": "Memory Search Fallback",
+  "agents.defaults.memorySearch.local.modelPath": "Local Embedding Model Path",
+  "agents.defaults.memorySearch.store.path": "Memory Search Index Path",
+  "agents.defaults.memorySearch.chunking.tokens": "Memory Chunk Tokens",
+  "agents.defaults.memorySearch.chunking.overlap":
+    "Memory Chunk Overlap Tokens",
+  "agents.defaults.memorySearch.sync.onSessionStart": "Index on Session Start",
+  "agents.defaults.memorySearch.sync.onSearch": "Index on Search (Lazy)",
+  "agents.defaults.memorySearch.sync.watch": "Watch Memory Files",
+  "agents.defaults.memorySearch.sync.watchDebounceMs":
+    "Memory Watch Debounce (ms)",
+  "agents.defaults.memorySearch.query.maxResults": "Memory Search Max Results",
+  "agents.defaults.memorySearch.query.minScore": "Memory Search Min Score",
   "auth.profiles": "Auth Profiles",
   "auth.order": "Auth Profile Order",
   "auth.cooldowns.billingBackoffHours": "Billing Backoff (hours)",
@@ -215,6 +232,20 @@ const FIELD_HELP: Record<string, string> = {
     "Failure window (hours) for backoff counters (default: 24).",
   "agents.defaults.models":
     "Configured model catalog (keys are full provider/model IDs).",
+  "agents.defaults.memorySearch":
+    "Vector search over MEMORY.md and memory/*.md (per-agent overrides supported).",
+  "agents.defaults.memorySearch.provider":
+    'Embedding provider ("openai" or "local").',
+  "agents.defaults.memorySearch.local.modelPath":
+    "Local GGUF model path or hf: URI (node-llama-cpp).",
+  "agents.defaults.memorySearch.fallback":
+    'Fallback to OpenAI when local embeddings fail ("openai" or "none").',
+  "agents.defaults.memorySearch.store.path":
+    "SQLite index path (default: ~/.clawdbot/memory/{agentId}.sqlite).",
+  "agents.defaults.memorySearch.sync.onSearch":
+    "Lazy sync: reindex on first search after a change.",
+  "agents.defaults.memorySearch.sync.watch":
+    "Watch memory files for changes (chokidar).",
   "plugins.enabled": "Enable plugin/extension loading (default: true).",
   "plugins.allow":
     "Optional allowlist of plugin ids; when set, only listed plugins load.",

src/config/types.ts

@@ -996,6 +996,47 @@ export type AgentToolsConfig = {
   };
 };
 
+export type MemorySearchConfig = {
+  /** Enable vector memory search (default: true). */
+  enabled?: boolean;
+  /** Embedding provider mode. */
+  provider?: "openai" | "local";
+  /** Fallback behavior when local embeddings fail. */
+  fallback?: "openai" | "none";
+  /** Embedding model id (remote) or alias (local). */
+  model?: string;
+  /** Local embedding settings (node-llama-cpp). */
+  local?: {
+    /** GGUF model path or hf: URI. */
+    modelPath?: string;
+    /** Optional cache directory for local models. */
+    modelCacheDir?: string;
+  };
+  /** Index storage configuration. */
+  store?: {
+    driver?: "sqlite";
+    path?: string;
+  };
+  /** Chunking configuration. */
+  chunking?: {
+    tokens?: number;
+    overlap?: number;
+  };
+  /** Sync behavior. */
+  sync?: {
+    onSessionStart?: boolean;
+    onSearch?: boolean;
+    watch?: boolean;
+    watchDebounceMs?: number;
+    intervalMinutes?: number;
+  };
+  /** Query behavior. */
+  query?: {
+    maxResults?: number;
+    minScore?: number;
+  };
+};
+
 export type ToolsConfig = {
   allow?: string[];
   deny?: string[];
@@ -1070,6 +1111,7 @@ export type AgentConfig = {
   workspace?: string;
   agentDir?: string;
   model?: string;
+  memorySearch?: MemorySearchConfig;
   /** Human-like delay between block replies for this agent. */
   humanDelay?: HumanDelayConfig;
   identity?: IdentityConfig;
@@ -1534,6 +1576,8 @@ export type AgentDefaultsConfig = {
   contextPruning?: AgentContextPruningConfig;
   /** Compaction tuning and pre-compaction memory flush behavior. */
   compaction?: AgentCompactionConfig;
+  /** Vector memory search configuration (per-agent overrides supported). */
+  memorySearch?: MemorySearchConfig;
   /** Default thinking level when no /think directive is present. */
   thinkingDefault?: "off" | "minimal" | "low" | "medium" | "high";
   /** Default verbose level when no /verbose directive is present. */

src/config/zod-schema.ts

@@ -867,6 +867,48 @@ const AgentToolsSchema = z
   })
   .optional();
 
+const MemorySearchSchema = z
+  .object({
+    enabled: z.boolean().optional(),
+    provider: z.union([z.literal("openai"), z.literal("local")]).optional(),
+    fallback: z.union([z.literal("openai"), z.literal("none")]).optional(),
+    model: z.string().optional(),
+    local: z
+      .object({
+        modelPath: z.string().optional(),
+        modelCacheDir: z.string().optional(),
+      })
+      .optional(),
+    store: z
+      .object({
+        driver: z.literal("sqlite").optional(),
+        path: z.string().optional(),
+      })
+      .optional(),
+    chunking: z
+      .object({
+        tokens: z.number().int().positive().optional(),
+        overlap: z.number().int().nonnegative().optional(),
+      })
+      .optional(),
+    sync: z
+      .object({
+        onSessionStart: z.boolean().optional(),
+        onSearch: z.boolean().optional(),
+        watch: z.boolean().optional(),
+        watchDebounceMs: z.number().int().nonnegative().optional(),
+        intervalMinutes: z.number().int().nonnegative().optional(),
+      })
+      .optional(),
+    query: z
+      .object({
+        maxResults: z.number().int().positive().optional(),
+        minScore: z.number().min(0).max(1).optional(),
+      })
+      .optional(),
+  })
+  .optional();
+
 const AgentEntrySchema = z.object({
   id: z.string(),
   default: z.boolean().optional(),
@@ -874,6 +916,7 @@ const AgentEntrySchema = z.object({
   workspace: z.string().optional(),
   agentDir: z.string().optional(),
   model: z.string().optional(),
+  memorySearch: MemorySearchSchema,
   humanDelay: HumanDelaySchema.optional(),
   identity: IdentitySchema,
   groupChat: GroupChatSchema,
@@ -1098,6 +1141,7 @@ const AgentDefaultsSchema = z
     userTimezone: z.string().optional(),
     contextTokens: z.number().int().positive().optional(),
     cliBackends: z.record(z.string(), CliBackendSchema).optional(),
+    memorySearch: MemorySearchSchema,
     contextPruning: z
       .object({
         mode: z