feat: enhance audit logging, add MCP settings, and production docker setup

- Implemented comprehensive audit logging for Tasks, Users, Settings, Goals, Labels, AI Chat, and Rewards.
- Added Admin UI for MCP Server settings and Audit Logs.
- Created docker-compose-production.yml with Traefik configuration.
- Fixed backend bugs (missing storage methods, route closure).
- Added Audit Logging Guidelines.

shared/schema.ts

@@ -57,6 +57,8 @@ export const tasks = pgTable("tasks", {
   labelId: varchar("label_id").references(() => labels.id),
   energyLevel: text("energy_level").default("medium"), // 'low' | 'medium' | 'high'
   estimatedDuration: integer("estimated_duration"), // in minutes
+  parentTaskId: varchar("parent_task_id").references((): any => tasks.id), // Self-reference for subtasks
+  startDate: timestamp("start_date"), // For multi-day tasks
   dependencies: text("dependencies").array(), // Array of task IDs
   userId: varchar("user_id").references(() => users.id), // Added for ownership
 });
@@ -108,6 +110,7 @@ export const insertLabelSchema = createInsertSchema(labels).omit({
 
 export const insertTaskSchema = createInsertSchema(tasks, {
   dueDate: z.coerce.date().nullable(),
+  startDate: z.coerce.date().nullable(),
 }).omit({
   id: true,
   userId: true, // We will set this server-side
@@ -250,3 +253,56 @@ export type InsertPasswordResetToken = z.infer<typeof insertPasswordResetTokenSc
 export type PasswordResetToken = typeof passwordResetTokens.$inferSelect;
 
 export type InsertUserReward = z.infer<typeof insertUserRewardSchema>;
+
+// AI Chat Tables
+export const conversations = pgTable("conversations", {
+  id: varchar("id").primaryKey().default(sql`gen_random_uuid()`),
+  userId: varchar("user_id").references(() => users.id).notNull(),
+  title: text("title").notNull(),
+  createdAt: timestamp("created_at").defaultNow(),
+  updatedAt: timestamp("updated_at").defaultNow(),
+});
+
+export const messages = pgTable("messages", {
+  id: varchar("id").primaryKey().default(sql`gen_random_uuid()`),
+  conversationId: varchar("conversation_id").references(() => conversations.id).notNull(),
+  role: text("role").notNull(), // 'user' | 'assistant' | 'system'
+  content: text("content").notNull(),
+  createdAt: timestamp("created_at").defaultNow(),
+});
+
+export const insertConversationSchema = createInsertSchema(conversations).omit({
+  id: true,
+  createdAt: true,
+  updatedAt: true,
+});
+
+export const insertMessageSchema = createInsertSchema(messages).omit({
+  id: true,
+  createdAt: true,
+});
+
+export type InsertConversation = z.infer<typeof insertConversationSchema>;
+export type Conversation = typeof conversations.$inferSelect;
+export type InsertMessage = z.infer<typeof insertMessageSchema>;
+export type Message = typeof messages.$inferSelect;
+
+// Audit Logging
+export const auditLogs = pgTable("audit_logs", {
+  id: varchar("id").primaryKey().default(sql`gen_random_uuid()`),
+  userId: varchar("user_id").references(() => users.id), // Nullable if system action (though usually we track actor)
+  action: text("action").notNull(), // 'CREATE', 'UPDATE', 'DELETE', 'LOGIN', etc.
+  entityType: text("entity_type").notNull(), // 'TASK', 'GOAL', 'USER', 'SYSTEM'
+  entityId: text("entity_id"), // ID of the modified entity
+  details: json("details"), // JSON object with changed fields
+  source: text("source").notNull().default("USER"), // 'USER' | 'AI' | 'SYSTEM'
+  createdAt: timestamp("created_at").defaultNow(),
+});
+
+export const insertAuditLogSchema = createInsertSchema(auditLogs).omit({
+  id: true,
+  createdAt: true,
+});
+
+export type InsertAuditLog = z.infer<typeof insertAuditLogSchema>;
+export type AuditLog = typeof auditLogs.$inferSelect;