added ability to use pro mode, to do complex timeseries/location searches

apps/cf-ai-backend/src/index.ts

@@ -407,6 +407,7 @@ app.post(
 			spaces: z.string().optional(),
 			sourcesOnly: z.string().optional().default("false"),
 			model: z.string().optional().default("gpt-4o"),
+			proMode: z.string().optional().default("false"),
 		}),
 	),
 	zValidator("json", chatObj),
@@ -415,6 +416,7 @@ app.post(
 		const body = c.req.valid("json");
 
 		const sourcesOnly = query.sourcesOnly === "true";
+		const proMode = query.proMode === "true";
 
 		// Return early for dumb requests
 		if (sourcesOnly && body.sources) {
@@ -422,7 +424,6 @@ app.post(
 		}
 
 		const spaces = query.spaces?.split(",") ?? [undefined];
-		console.log(spaces);
 
 		// Get the AI model maker and vector store
 		const { model, store } = await initQuery(c, query.model);
@@ -431,14 +432,67 @@ app.post(
 			const filter: VectorizeVectorMetadataFilter = {
 				[`user-${query.user}`]: 1,
 			};
-			console.log("Spaces", spaces);
+
+			let proModeListedQueries: string[] = [];
+
+			if (proMode) {
+				const addedToQuery = (await c.env.AI.run(
+					// @ts-ignore
+					"@hf/nousresearch/hermes-2-pro-mistral-7b",
+					{
+						messages: [
+							{
+								role: "system",
+								content:
+									"You are a query enhancer. You must enhance a user's query to make it more relevant to what the user might be looking for. If there's any mention of dates like 'last summer' or 'this year', you should return 'DAY: X, MONTH: Y, YEAR: Z'. If there's any mention of locations, add that to the query too. Try to keep your responses as short as possible. Add to the user's query, don't replace it. Make sure to keep your answers short.",
+							},
+							{ role: "user", content: query.query },
+						],
+						tools: [
+							{
+								type: "function",
+								function: {
+									name: "Enhance query get list",
+									description:
+										"Enhance the user's query to make it more relevant",
+									parameters: {
+										type: "object",
+										properties: {
+											listedQueries: {
+												type: "array",
+												description: "List of queries that the user has asked",
+												items: {
+													type: "string",
+												},
+											},
+										},
+										required: ["Enhance query get list"],
+									},
+								},
+							},
+						],
+						max_tokens: 200,
+					},
+				)) as {
+					response?: string;
+					tool_calls?: {
+						name: string;
+						arguments: {
+							listedQueries: string[];
+						};
+					}[];
+				};
+
+				proModeListedQueries =
+					addedToQuery.tool_calls?.[0]?.arguments?.listedQueries ?? [];
+			}
 
 			// Converting the query to a vector so that we can search for similar vectors
-			const queryAsVector = await store.embeddings.embedQuery(query.query);
+			const queryAsVector = await store.embeddings.embedQuery(
+				query.query + " " + proModeListedQueries.join(" "),
+			);
 			const responses: VectorizeMatches = { matches: [], count: 0 };
 
-			console.log("hello world", spaces);
-
 			// SLICED to 5 to avoid too many queries
 			for (const space of spaces.slice(0, 5)) {
 				if (space && space.length >= 1) {
@@ -506,7 +560,12 @@ app.post(
 
 				const metadata = normalizedData.map((datapoint) => datapoint.metadata);
 
-				return c.json({ ids: storedContent, metadata, normalizedData });
+				return c.json({
+					ids: storedContent,
+					metadata,
+					normalizedData,
+					proModeListedQueries,
+				});
 			}
 		}
 

apps/cf-ai-backend/src/utils/OpenAIEmbedder.ts

@@ -50,7 +50,7 @@ export class OpenAIEmbeddings {
 		const json = zodTypeExpected.safeParse(data);
 
 		if (!json.success) {
-			console.log(data);
+			console.log(JSON.stringify(data));
 			throw new Error("Invalid response from OpenAI: " + json.error.message);
 		}
 

apps/web/app/(dash)/chat/[chatid]/loading.tsx

@@ -7,9 +7,15 @@ async function Page({
 }: {
 	searchParams: Record<string, string | string[] | undefined>;
 }) {
-	const q = (searchParams?.q as string) ?? "from_loading";
+	const q = (searchParams?.q as string) ?? "";
 	return (
-		<ChatWindow q={q} spaces={[]} initialChat={undefined} threadId={"idk"} />
+		<ChatWindow
+			proMode={false}
+			q={q}
+			spaces={[]}
+			initialChat={undefined}
+			threadId={"idk"}
+		/>
 	);
 }
 

apps/web/app/(dash)/chat/[chatid]/page.tsx

@@ -9,7 +9,8 @@ async function Page({
 	params: { chatid: string };
 	searchParams: Record<string, string | string[] | undefined>;
 }) {
-	const { firstTime, q, spaces } = chatSearchParamsCache.parse(searchParams);
+	const { firstTime, q, spaces, proMode } =
+		chatSearchParamsCache.parse(searchParams);
 
 	let chat: Awaited<ReturnType<typeof getFullChatThread>>;
 
@@ -31,6 +32,7 @@ async function Page({
 			spaces={spaces ?? []}
 			initialChat={chat.data.length > 0 ? chat.data : undefined}
 			threadId={params.chatid}
+			proMode={proMode}
 		/>
 	);
 }

apps/web/app/(dash)/chat/chatWindow.tsx

@@ -35,14 +35,19 @@ function ChatWindow({
 				parts: [],
 				sources: [],
 			},
+			proModeProcessing: {
+				queries: [],
+			},
 		},
 	],
 	threadId,
+	proMode,
 }: {
 	q: string;
 	spaces: { id: number; name: string }[];
 	initialChat?: ChatHistory[];
 	threadId: string;
+	proMode: boolean;
 }) {
 	const [layout, setLayout] = useState<"chat" | "initial">("chat");
 	const [chatHistory, setChatHistory] = useState<ChatHistory[]>(initialChat);
@@ -63,13 +68,17 @@ function ChatWindow({
 
 	const router = useRouter();
 
-	const getAnswer = async (query: string, spaces: string[]) => {
+	const getAnswer = async (
+		query: string,
+		spaces: string[],
+		proMode: boolean = false,
+	) => {
 		if (query.trim() === "from_loading" || query.trim().length === 0) {
 			return;
 		}
 
 		const sourcesFetch = await fetch(
-			`/api/chat?q=${query}&spaces=${spaces}&sourcesOnly=true&threadId=${threadId}`,
+			`/api/chat?q=${query}&spaces=${spaces}&sourcesOnly=true&threadId=${threadId}&proMode=${proMode}`,
 			{
 				method: "POST",
 				body: JSON.stringify({ chatHistory }),
@@ -91,6 +100,8 @@ function ChatWindow({
 			behavior: "smooth",
 		});
 
+		let proModeListedQueries: string[] = [];
+
 		const updateChatHistoryAndFetch = async () => {
 			// Step 1: Update chat history with the assistant's response
 			await new Promise((resolve) => {
@@ -123,6 +134,11 @@ function ChatWindow({
 						).length,
 					}));
 
+					lastAnswer.proModeProcessing.queries =
+						sourcesParsed.data.proModeListedQueries ?? [];
+
+					proModeListedQueries = lastAnswer.proModeProcessing.queries;
+
 					resolve(newChatHistory);
 					return newChatHistory;
 				});
@@ -130,7 +146,7 @@ function ChatWindow({
 
 			// Step 2: Fetch data from the API
 			const resp = await fetch(
-				`/api/chat?q=${query}&spaces=${spaces}&threadId=${threadId}`,
+				`/api/chat?q=${(query += proModeListedQueries.join(" "))}&spaces=${spaces}&threadId=${threadId}`,
 				{
 					method: "POST",
 					body: JSON.stringify({ chatHistory, sources: sourcesParsed.data }),
@@ -181,6 +197,7 @@ function ChatWindow({
 				getAnswer(
 					q,
 					spaces.map((s) => `${s.id}`),
+					proMode,
 				);
 			}
 		} else {
@@ -224,6 +241,28 @@ function ChatWindow({
 											{chat.question}
 										</h2>
 
+										{chat.proModeProcessing?.queries?.length > 0 && (
+											<div className="flex flex-col mt-2">
+												<div className="text-foreground-menu py-2">
+													Pro Mode
+												</div>
+												<div className="text-base">
+													<div className="flex gap-2 text-base">
+														{chat.proModeProcessing.queries.map(
+															(query, idx) => (
+																<div
+																	className="bg-secondary rounded-md p-2"
+																	key={`promode-query-${idx}`}
+																>
+																	{query}
+																</div>
+															),
+														)}
+													</div>
+												</div>
+											</div>
+										)}
+
 										<div className="flex flex-col mt-2">
 											<div>
 												<div className="text-foreground-menu py-2">Answer</div>
@@ -407,6 +446,9 @@ function ChatWindow({
 													parts: [],
 													sources: [],
 												},
+												proModeProcessing: {
+													queries: [],
+												},
 											},
 										];
 									});

apps/web/app/(dash)/header/autoBreadCrumbs.tsx

@@ -13,8 +13,6 @@ import React from "react";
 function AutoBreadCrumbs() {
 	const pathname = usePathname();
 
-	console.log(pathname.split("/").filter(Boolean));
-
 	return (
 		<Breadcrumb className="hidden md:block">
 			<BreadcrumbList>
@@ -31,7 +29,7 @@ function AutoBreadCrumbs() {
 					.filter(Boolean)
 					.map((path, idx, paths) => (
 						<>
-							<BreadcrumbItem key={path}>
+							<BreadcrumbItem key={path + idx}>
 								<BreadcrumbLink href={`/${paths.slice(0, idx + 1).join("/")}`}>
 									{path.charAt(0).toUpperCase() + path.slice(1)}
 								</BreadcrumbLink>

apps/web/app/(dash)/home/history.tsx

@@ -26,9 +26,18 @@ const History = memo(() => {
 		<ul className="text-base list-none space-y-3 text-[#b9b9b9] mt-8">
 			{!chatThreads_ && (
 				<>
-					<Skeleton className="w-[80%] h-4 bg-[#3b444b] "></Skeleton>
-					<Skeleton className="w-[40%] h-4 bg-[#3b444b] "></Skeleton>
-					<Skeleton className="w-[60%] h-4 bg-[#3b444b] "></Skeleton>
+					<Skeleton
+						key="loader-1"
+						className="w-[80%] h-4 bg-[#3b444b] "
+					></Skeleton>
+					<Skeleton
+						key="loader-2"
+						className="w-[40%] h-4 bg-[#3b444b] "
+					></Skeleton>
+					<Skeleton
+						key="loader-3"
+						className="w-[60%] h-4 bg-[#3b444b] "
+					></Skeleton>
 				</>
 			)}
 			{chatThreads_?.map((thread) => (
@@ -36,6 +45,7 @@ const History = memo(() => {
 					initial={{ opacity: 0, filter: "blur(1px)" }}
 					animate={{ opacity: 1, filter: "blur(0px)" }}
 					className="flex items-center gap-2 truncate"
+					key={thread.id}
 				>
 					<ArrowLongRightIcon className="h-5" />{" "}
 					<Link prefetch={false} href={`/chat/${thread.id}`}>

apps/web/app/(dash)/home/page.tsx

@@ -103,13 +103,12 @@ function Page({ searchParams }: { searchParams: Record<string, string> }) {
 				<QueryInput
 					initialQuery={query}
 					setQueryPresent={setQueryPresent}
-					handleSubmit={async (q, spaces) => {
+					handleSubmit={async (q, spaces, proMode) => {
 						if (q.length === 0) {
 							toast.error("Query is required");
 							return;
 						}
 
-						console.log("creating thread");
 						const threadid = await createChatThread(q);
 
 						if (!threadid.success || !threadid.data) {
@@ -117,9 +116,8 @@ function Page({ searchParams }: { searchParams: Record<string, string> }) {
 							return;
 						}
 
-						console.log("pushing to chat");
 						push(
-							`/chat/${threadid.data}?spaces=${JSON.stringify(spaces)}&q=${q}`,
+							`/chat/${threadid.data}?spaces=${JSON.stringify(spaces)}&q=${q}&proMode=${proMode}`,
 						);
 					}}
 					initialSpaces={spaces}

apps/web/app/(dash)/home/queryinput.tsx

@@ -1,6 +1,6 @@
 "use client";
 
-import React, { useState } from "react";
+import React, { useEffect, useState } from "react";
 import { FilterSpaces } from "./filterSpaces";
 import { ArrowRightIcon } from "@repo/ui/icons";
 import Image from "next/image";
@@ -20,7 +20,11 @@ function QueryInput({
 	}[];
 	initialQuery?: string;
 	mini?: boolean;
-	handleSubmit: (q: string, spaces: { id: number; name: string }[]) => void;
+	handleSubmit: (
+		q: string,
+		spaces: { id: number; name: string }[],
+		proMode: boolean,
+	) => void;
 }) {
 	const [q, setQ] = useState(initialQuery || "");
 
@@ -41,7 +45,7 @@ function QueryInput({
 						if (q.trim().length === 0) {
 							return;
 						}
-						handleSubmit(q, selectedSpaces);
+						handleSubmit(q, selectedSpaces, proMode);
 						setQ("");
 					}}
 				>
@@ -58,7 +62,7 @@ function QueryInput({
 								if (q.trim().length === 0) {
 									return;
 								}
-								handleSubmit(q, selectedSpaces);
+								handleSubmit(q, selectedSpaces, proMode);
 								setQ("");
 							}
 						}}
@@ -83,7 +87,7 @@ function QueryInput({
 								</Label>
 								<Switch
 									value={proMode ? "on" : "off"}
-									onChange={() => setProMode((prev) => !prev)}
+									onCheckedChange={(v) => setProMode(v)}
 									id="pro-mode"
 									about="Pro mode"
 								/>

apps/web/app/api/chat/route.ts

@@ -27,6 +27,7 @@ export async function POST(req: NextRequest) {
 	const spaces = url.searchParams.get("spaces");
 
 	const sourcesOnly = url.searchParams.get("sourcesOnly") ?? "false";
+	const proMode = url.searchParams.get("proMode") === "true";
 
 	const jsonRequest = (await req.json()) as {
 		chatHistory: ChatHistory[];
@@ -55,7 +56,7 @@ export async function POST(req: NextRequest) {
 	const modelCompatible = await convertChatHistoryList(validated.data);
 
 	const resp = await fetch(
-		`${process.env.BACKEND_BASE_URL}/api/chat?query=${query}&user=${session.user.id}&sourcesOnly=${sourcesOnly}&spaces=${spaces}`,
+		`${process.env.BACKEND_BASE_URL}/api/chat?query=${query}&user=${session.user.id}&sourcesOnly=${sourcesOnly}&spaces=${spaces}&proMode=${proMode}`,
 		{
 			headers: {
 				Authorization: `Bearer ${process.env.BACKEND_SECURITY_KEY}`,

apps/web/lib/searchParams.ts

@@ -32,4 +32,5 @@ export const chatSearchParamsCache = createSearchParamsCache({
 
 		return valid.data;
 	}),
+	proMode: parseAsBoolean.withDefault(false),
 });

packages/shared-types/index.ts

@@ -17,6 +17,9 @@ export const ChatHistoryZod = z.object({
 		sources: z.array(SourceZod),
 		justification: z.string().optional(),
 	}),
+	proModeProcessing: z.object({
+		queries: z.array(z.string()),
+	}),
 });
 
 export type ChatHistory = z.infer<typeof ChatHistoryZod>;
@@ -77,6 +80,7 @@ export const sourcesZod = z.object({
 	ids: z.array(z.string()),
 	metadata: z.array(z.any()),
 	normalizedData: z.array(z.any()).optional(),
+	proModeListedQueries: z.array(z.string()).optional(),
 });
 
 export type SourcesFromApi = z.infer<typeof sourcesZod>;