thecookingsenpai / autogpt-gui

A graphical user interface for AutoGPT

• 👋 TheCookingSenpai - Blockdrops LLC

• Fullstack developer

• Blockchain developer

Tech Used:

• Python

• JS (nodeJS, expressJS, parcelJS)

• Solidity

• Bash scripting

Areas of interest:

• NFT & Tokens

• Security

• New paradigm tools

• General purpose tools

• Generic forks

Contact:

• 📫 TG: @tcsenpai (https://t.me/tcsenpai)

Articles Written:

Blockchains and Security

• https://aiternalex.com/artificial-intelligence/blockchains-security/

autogpt-gui's Issues

[feature request] local http api

Request to utilize oogabooga as a local http api to make this system self reliant and fully local as far as processing is concerned.

Data lookup still being unrestricted to the web.

Nothing happening when I click Go!

The personality, name, and at least one (I've tried every number up to 5) goal is filled in. Have also tried checking the different boxes below to no avail.

it says "Please run: python -m autogpt"

when press "Go", it says "Please run: python -m autogpt".

i have check the .script/main.py and there is the content:

  from colorama import Style, init
  
  # Initialize colorama
  init(autoreset=True)
  
  # Use the bold ANSI style
  print(
      f"""{Style.BRIGHT}Please run:
  python -m autogpt
  """
)

i think the lastest verison is not allowed the run the cmd by

let args_cmd = ['scripts/main.py']

python: can't open file '/Users/maximiliandoelle/Projects/Auto-GPT/scripts/main.py': [Errno 2] No such file or directory

Seems like it is looking in the wrong place - the main.py script is in the root directory

can you add input mode?

To start this project, you need to do the following:

1.Modify script.js: Replace let args_cmd = ['main.py'] with let args_cmd = ['-m', 'autogpt'].

2.You can choose not to use the continuous method: Open the developer tools option and enter whether you want to continue or stop. Then, use :
child_autogpt.stdin.write('y\n');

I hope the author can improve this project as soon as possible.

FATAL:credentials.cc(127)] Check failed: . : Permission denied (13)

can not startup ，when exec the command npm start ，it appears the following error

npm start

[email protected] start
electron .

[26406:0612/201648.101773:FATAL:credentials.cc(127)] Check failed: . : Permission denied (13)
/home/xxx/Auto-GPT/autogpt-gui/node_modules/electron/dist/electron exited with signal SIGTRAP

Block Seller Product Form Submission Till Video Uploading Completes

Currently, the seller is unaware of the video uploading progress. The task is to track & display the video uploading progress to the seller and ensure that he is able to submit the form ONLY WHEN THE VIDEO('S) HAVE BEEN UPLOADED SUCCESSFULLY!

Autocelopt

AutoGPT has been updated, GUI broken, I fixed script.js with GPT4. Just need add auto-gpt.Json from old version into AutoGPT for warning removal. Sorry if this is not correct place, delete as necessary..

const fs = require('fs');
const path = require('path')
const { spawn } = require('child_process');
const { error } = require('console');
var child_autogpt = null
var first_input = false
let save_output = false

const cmd_execute = document.getElementById('cmd_execute')
const cmd_stop = document.getElementById('cmd_stop')
const cfg_debug = document.getElementById('cfg_debug')
const cfg_continuous = document.getElementById('cfg_continuous')
const cfg_speak = document.getElementById('cfg_speak')
const cfg_save = document.getElementById('cfg_save')
const util_installDeps = document.getElementById('util_installDeps')
const zone_personality = document.getElementById('zone_personality')
const zone_name = document.getElementById('zone_name')
const zone_output = document.getElementById('zone_output')
const zone_error = document.getElementById('zone_error')

var is_running = false

async function init() {
cmd_execute.addEventListener('click', async () => {
if (is_running) return
zone_output.innerHTML = ''
let args = await get_args_from_gui()
is_running = true
await execute_autogpt(args)
is_running = false
});

cmd_stop.addEventListener('click', async () => {
    if (child_autogpt) {
        child_autogpt.kill()
        cmd_stop.classList.add('disabled')
        zone_output.scrollTop = zone_output.scrollHeight
    }
});

}

async function get_args_from_gui() {
let _args = {
name: zone_name.value,
personality: zone_personality.value,
goals: [],
debug: cfg_debug.checked,
continuous: cfg_continuous.checked,
speak: cfg_speak.checked,
save: cfg_save.checked
}
for (let i = 1; i <= 5; i++) {
let goal = document.getElementById(zone_goal_${i}).value
_args.goals.push(goal)
}
return _args
}

async function execute_autogpt(args) {
if (!zone_error.classList.contains('hidden')) {
zone_error.classList.add('hidden')
}
zone_error.innerHTML = ''
if (args.goals[0] == '') {
if (zone_error.classList.contains('hidden')) zone_error.classList.remove('hidden')
zone_error.innerHTML = 'You need to specify at least one goal'
return
}
if (args.personality == '') {
if (zone_error.classList.contains('hidden')) zone_error.classList.remove('hidden')
zone_error.innerHTML = 'You need to specify a personality'
return
}
if (args.name == '') {
args.name = 'Autogpt'
}
let settings = "ai_goals: \n"
for (let i = 0; i < args.goals.length; i++) {
settings += - ${args.goals[i]}\n
}
settings += "ai_name: " + args.name + "\n"
settings += "ai_role: " + args.personality + "\n"
fs.writeFileSync('../ai_settings.yaml', settings)
let additionals = []
if (args.debug) {
additionals.push('--debug')
}
if (args.continuous) {
additionals.push('--continuous')
}
if (args.speak) {
additionals.push('--speak')
}
if (args.save) {
save_output = true
}
let cmd = 'python'
let args_cmd = ['-m', 'autogpt']
args_cmd.push(...additionals)
child_autogpt = spawn(cmd, args_cmd, {cwd: path.join(__dirname, '..')})

if (child_autogpt.pid) {
    cmd_stop.classList.remove('hidden')
}

child_autogpt.stdout.on('data', (data) => {
    if (data.includes('Thinking...')) {
        if (zone_output.innerHTML.includes('Thinking...') ){
            data = "."
        }
    }
    if (data.includes('Thinking...')) {
        zone_output.innerHTML += '<br>'
    }
    if (data.includes('\n')) {
        zone_output.innerHTML += '<br>|>|> '
    }
    zone_output.innerHTML += data
    zone_output.scrollTop = zone_output.scrollHeight

    if (save_output) {
        fs.appendFileSync(path.join(__dirname, 'output.txt'), data)
    }
    if (data.includes('(y/n)')) {
        if (!first_input) {
            first_input = true
            child_autogpt.stdin.write('y\n')
        }
    }
});

child_autogpt.stderr.on('data', (data) => {
    zone_output.innerHTML += '<br><span class"red">' + data + '</span>'
});

child_autogpt.on('close', (code) => {
    zone_output.innerHTML += '<br><p class"red">child process exited with code ' + code + '</p>'
    cmd_stop.classList.add('hidden')
});

}

(async () => {
await init()
}) ();

started running but then couldn't find the scripts direcotry - does this work with the docker version of autoGPT?

python: can't open file 'C:\Auto-GPT\scripts\main.py': [Errno 2] No such file or directory
child process exited with code 2

一起研究 +V TB-520BB

一起研究 +V TB-520BB

[2448:0525/115511.103:ERROR:gpu_init.cc(523)] Passthrough is not supported, GL is disabled, ANGLE is

The search results you provided do not contain any relevant information for the given query. However, based on the command "npm install" and the subsequent output, it appears to be related to the npm (Node Package Manager) tool used for managing dependencies in Node.js projects.

When you run the command "npm install," it installs the packages specified in the project's package.json file. In this case, it added 72 packages and audited 73 packages in 1 minute [1]. Additionally, it indicates that 18 packages are looking for funding and suggests running "npm fund" for details [1].

The output also states that no vulnerabilities were found in the installed packages [1]. However, it notifies that there is a new patch version of npm available (9.6.7) and provides a link to the changelog for version 9.6.7 [1]. To update npm to the new version, you can run the command "npm install -g [email protected]" [1].

Regarding the command "npm start," it appears to be related to running an Electron application. The output shows that the electron-quick-start project is being started with the command "electron ." [2]. However, the provided search results do not provide any further information about this specific command.

To summarize, the command "npm install" is used to install packages in a Node.js project, and it installs the specified packages while also auditing them for vulnerabilities. Running "npm start" can be used to start an Electron application. For further details on the npm commands or specific packages, additional search queries or information would be needed.

References:
[1] npm output: https://github.com/npm/cli/releases/tag/v9.6.7
[2] electron-quick-start: https://github.com/electron/electron-quick-start

Ayuda

Confused about the blank Your AI is?

In the first blank, it said
Your AI is
what should I input here? like Davinci or chatgpt3.5?

Have you faced the problem: electron: error while loading shared libraries: libatk-1.0.so.0?

[datai]$ npm start

[email protected] start /data/Auto-GPT/autogpt-gui
electron .

/data/Auto-GPT/autogpt-gui/node_modules/electron/dist/electron: error while loading shared libraries: libatk-1.0.so.0: cannot open shared object file: No such file or directory
npm ERR! code ELIFECYCLE
npm ERR! syscall spawn
npm ERR! file sh
npm ERR! errno ENOENT
npm ERR! [email protected] start: electron .
npm ERR! spawn ENOENT
npm ERR!
npm ERR! Failed at the [email protected] start script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

npm ERR! A complete log of this run can be found in:
npm ERR! /home/data/.npm/_logs/2023-05-26T06_33_08_359Z-debug.log

Does anyone know how could I fix this problem? thanks

Unable to confirm action

Am unable to confirm action after "Enter 'y' to authorise command, 'y -N' to run N continuous commands, 'n' to exit program, or enter feedback for"

Anti-Corruption Unit