I'm having trouble with a bad gateway error. What configuration do I need to set up?

Hey I'm running this behind a cloudflared tunnel, and I'm getting a 401 unauthorized when sending any message, however trying the service locally on the same server works fine.

Is there a way to enable a debug mode to see what the issue is?

The issue is that the app is just hardcoded to use http://localhost for everything internal.

with all default settings, freshly cloned repo, the app goes into redirect loop after logging in.

Environment:

• macOS: Sonoma 14.3.1 (23D60)
• node version: v20.12.1
• Browser: firefox and safari
• commit: 5bb59c3

Hello, I just cloned the repo in a fresh Debian 12 container with node v21.7.3.

upon running npm run start i got the following error:

> [email protected] start
> node server.js

/home/user/GPTPortal/server.js:12
app.use(bodyParser.json({ limit: '50mb' }));
^

ReferenceError: Cannot access 'app' before initialization
    at Object.<anonymous> (/home/giorgio/GPTPortal/server.js:12:1)
    at Module._compile (node:internal/modules/cjs/loader:1368:14)
    at Module._extensions..js (node:internal/modules/cjs/loader:1426:10)
    at Module.load (node:internal/modules/cjs/loader:1205:32)
    at Module._load (node:internal/modules/cjs/loader:1021:12)
    at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:142:12)
    at node:internal/main/run_main_module:28:49

Which i solved by rearranging some lines in the server.js file:

// importing required node packages
require('dotenv').config();
const express = require('express');
const axios = require('axios');
const basicAuth = require('express-basic-auth');
const fs = require('fs');
const { marked } = require('marked');
const bodyParser = require('body-parser');
const app = express(); // Move this line up here
const download = require('image-downloader');
const cors = require('cors');
// Increase the limit for JSON bodies
app.use(bodyParser.json({ limit: '50mb' }));
app.use(bodyParser.urlencoded({ limit: '50mb', extended: true, parameterLimit: 50000 }));
app.use(express.json()); // for parsing application/json
app.use(express.static('public')); // Serves your static files from 'public' directory
app.use(cors());
const router = express.Router();
const { v4: uuidv4 } = require('uuid');

// openai
const OpenAI = require('openai').default;
// <other code remains the same>

However, when i try to access the application from the network: http://<local-server-ip>:3000/portal i get into a redirect loop to http://<local-server-ip>:3000/public/portal.html The page isn’t redirecting properly.

I'd appreciate some ideas on how to fix this, i'd love to try this app.

Hello! Thanks for your work! Is it possible to add stream=true parameter to have streamed responses?

Latest docker version when typing a message always returns "Error sending message. Please try again." Attaching yaml file and screenshot
yaml.txt

Please add Screenshots to your README.md .

People are visual and want to know what they get.

https://gpt-portal.com/ doesn not seem to be your website and you have no links to a website.

So help interested users out to see what they get.

A docker-compose.yml is available which is great!

Greetings!

I must be missing something.
npm install runs fine
copied .env.example to .env
I configure the .env with my keys and username and password
changed nothing else
start the server.js
load localhost:3000 in firefox or chrome
http://localhost:3000/portal redirects to http://localhost:3000/public/portal.html

get a redirect loop

running on windows 11,

node --version
v20.9.0

any suggestions?

Thanks

I set the default model as gemini-1.5-flash and I already choose it from the dropdown menu but when I ask send the prompt I get this error message:

Error with Gemini API: [GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1/models/gemini-1.5-flash:generateContent: fetch is not defined

Here's the web console log:
Error sending message to server: Error: HTTP error! status: 500
at sendMessageToServer (script.js:2198:19)
at async HTMLButtonElement. (script.js:1848:15)

Running the app on Windows 11.

First, thanks for the great project!
It would be great if you could add support for the assistant API, which allows users to use the custom assistants they created on the dev platform.
Assistants unlike normal ChatGPT aren't just custom prompts, they can read files like CSVs, which can't be easily replicated with a system prompt.

https://platform.openai.com/docs/assistants/overview

Hi, great app. Congrats bro. I was just wondering if you have thought in updating the app to use the new Assistants API which is better to use with custom instructions and AI roles.

https://platform.openai.com/docs/assistants/overview?lang=node.js

Hello,

I use another program that allows to customize "API Proxy Address", for OpenAI and Anthropic.

They also allow custom AI endpoints, if they are OpenIA compatible.

Do you plan to do the same.

Thanks

Any idea? MacOS 14

TypeError [ERR_INVALID_ARG_TYPE]: The "string" argument must be of type string or an instance of Buffer or ArrayBuffer. Received undefined
    at new NodeError (node:internal/errors:405:5)
    at Function.byteLength (node:buffer:787:11)
    at safeCompare (/Users/[user[Downloads/GPTPortal-main/node_modules/express-basic-auth/index.js:9:33)
    at staticUsersAuthorizer (/Users/USER/Downloads/GPTPortal-main/node_modules/express-basic-auth/index.js:42:43)
    at authMiddleware (/Users/USER/Downloads/GPTPortal-main/node_modules/express-basic-auth/index.js:61:18)
    at Layer.handle [as handle_request] (/Users/m1/Downloads/GPTPortal-main/node_modules/express/lib/router/layer.js:95:5)
    at trim_prefix (/Users/USER/Downloads/GPTPortal-main/node_modules/express/lib/router/index.js:328:13)
    at /Users/m1/Downloads/GPTPortal-main/node_modules/express/lib/router/index.js:286:9
    at Function.process_params (/Users/m1/Downloads/GPTPortal-main/node_modules/express/lib/router/index.js:346:12)
    at next (/Users/USER/Downloads/GPTPortal-main/node_modules/express/lib/router/index.js:280:10)