miurla / babyagi-ui Goto Github PK

Hello, I'm just trying BabyGPT right now but I'm having this error when launching a task :
'Request failed with status code 429'

It seems correlated with too musch requests in some ways.

I have not set a PINECONE_API_KEY, just a valid OpenAI key inside the UI.

I'm searching but I can't figured it out

• GPT-4 alert
• Supplementation of babyagi

Reference

• Twitter: https://twitter.com/yoheinakajima/status/1666313838868992001
• GitHub: https://github.com/yoheinakajima/babyagi/blob/main/classic/BabyDeerAGI.py

TODO

• Putting together a flow
• Add Executer
• Add a menu
• Implement Execution
• Tools
  • Update web-search
  • Addition of user-input
• Parallel execution support

ref: #113

hi,
I updated last version yesterday
But, i realize errors :

With google chrome:
First : in cmd when in use : "Fast Refresh had to perform a full reload"
Second : New IntroGuide.tsx] menu is not visible in chrome but sidebar menu with istoric is okay
With Firefox:
First : No problem in cmdlog : No "Fast Refresh had to perform a full reload" anymore
Second: New [IntroGuide.tsx] menu is visible but sidebar menu with historic text is not visible

any conflit UI ?
Thanx do reporting solution

AbyssTensor

Is there a way to access files it said it created?

Currently available only by client request.

https://vercel.com/blog/introducing-the-vercel-ai-sdk
This looks good to use.

Error Message - using Vercel for deployment.

I updated the .env file and entered a working API key for SerpAPI (I know it works since I'm using it for another project). Did I miss somewhere else to configure Serp?

{"message":"Request failed with status code 401","name":"AxiosError","stack":"AxiosError: Request failed with status code 401\n at settle (file:///var/task/node_modules/axios/lib/core/settle.js:19:12)\n at IncomingMessage.handleStreamEnd (file:///var/task/node_modules/axios/lib/adapters/http.js:570:11)\n at IncomingMessage.emit (node:events:525:35)\n at endReadableNT (node:internal/streams/readable:1359:12)\n at process.processTicksAndRejections (node:internal/process/task_queues:82:21)","config":{"transitional":{"silentJSONParsing":true,"forcedJSONParsing":true,"clarifyTimeoutError":false},"adapter":["xhr","http"],"transformRequest":[null],"transformResponse":[null],"timeout":0,"xsrfCookieName":"XSRF-TOKEN","xsrfHeaderName":"X-XSRF-TOKEN","maxContentLength":-1,"maxBodyLength":-1,"env":{},"headers":{"Accept":"application/json, text/plain, /","User-Agent":"axios/1.4.0","Accept-Encoding":"gzip, compress, deflate, br"},"params":{"engine":"google","q":"Research the impact of Generative AI on Retail/eCommerce","num":5},"method":"get","url":"https://serpapi.com/search"},"code":"ERR_BAD_REQUEST","status":401}

Save the execution results and display the history in a list.
ref: #9

UI

• Sidebar
  • Button (New)
  • List

Functions

• Add data with the new button
• Add data with the start button
• Save data when execution is finished
• Display history
• Delete data

Data

• id: string
• title: string
• date: Date()
• params
  • mode id: string
  • iterations: number
  • first_task: string
  • objective: string
• messages: Message[]

Save destination

• LocalStorage

Crear_4_sprints_para_ser_montado_en_Azure_Devops_Cada_sprint_con_10_dias_habiles_o_2_semanas_calendario_Listar_actividades_y_tareas_por_sprint_Cada_actividad_y_tarea_con_numero_de_horas_de_esfuerzo_para_dos_desar.txt

and markdown to:

How can contribute or collaborate for translate corrrectly to spanish?

Allow Agent code to be generated from npm commands

Final deliverables 📁

• {agent_name}
  • {agent_name} Agent.ts
  • {agent_name} Prompt.ts

reference: https://github.com/miurla/babyagi-ui/tree/main/src/agents/babyagiui-mod/agents/taskCompletion

| ID | Status | Task | Tool | Dependency |

Failed to generate text.

All tasks completed!

That some strings in web page and txt downloaded not translated to spanish
Tell me where and how can PR with translated strings

This prompt appeared when I just set it up and started running. I had no idea what was wrong. Please help me

Consider data recovery options, such as logging to local storage.

Error handling as we are doing miscellaneous implementation for start-up.

When running BabyAgi-UI, noticed in the terminal providing this error.

wait  - compiling /api/enrich (client and server)...
event - compiled successfully in 48 ms (280 modules)
error - [RangeError: Invalid status code: 0] {
  code: 'ERR_HTTP_INVALID_STATUS_CODE'

Than later on I may get a error like this

error - [RangeError: Invalid status code: 0] {
  code: 'ERR_HTTP_INVALID_STATUS_CODE'
}

Any clarification if this is a major issue or not would be greatly appreciated, also is there way for it to continue or save its previous sessions?

Since it is still in the execution log, return the URL in messageCallback.

Screenshot

Hey guys!

• I follow the Getting Started tutorial : https://github.com/miurla/babyagi-ui#-getting-started

• I don't use Pinecone index.

• I filled the OpenAI API key from the web interface.

System Configuration :

• OS : Ventura 13.3.1 (a)
• Chip : Apple M1 Pro

BabyAGI configuration :

• Model : OpenAI gpt-4
• Agent : BabyCatAGI

When I send a new objective, I got this error.

Thank you for your good work @miurla

Getting this error after inputting my objective and hit the arrow button
Unhandled Runtime Error
TypeError: Cannot read properties of undefined (reading 'trim')

error - [RangeError: Invalid status code: 0] {
code: 'ERR_HTTP_INVALID_STATUS_CODE'

That's the question, I can't use OpenAI and I would love to run this BabyAGI over the GPU in my local computer with some models like WizardLM or Gpt4-x-Vicuna, both quantized.

Do you plan to make a local version of this?

Thanks for this!

Be able to receive evaluation and logging of results from LLM, like ChatGPT

Image

Enhancing the Project with Suggestions from Godmode-GPT https://godmode.space/

Hi @miurla,

First of all, I would like to commend your efforts on adding internet access and scraping capabilities to the project. Your hard work has been exceptional, and I'd like to offer some suggestions to further enhance your work.

I have based my observations on the advantages provided by the Godmode-GPT project (AutoGPT+BabyAGI combo):

1. Editable Initial Task List: Consider making the initial task list generated by GPT-4 editable or reprioritizable by humans, which will help BabyAGI narrow down to the actual problem more efficiently.

2. Global Context Maintenance: In the current implementation, once iterations or tasks are finished, the next human input is treated as a new task. If you could integrate the human input into the current context and continue within that context, more effective solutions may be discovered.

3. Local File Storage: To avoid experiences like @sacredgrove23's, consider storing intermediate results in text files within local storage. This would require a file explorer in the UI and compartmentalization of files among chats, which may be a complex task, but it would enhance the overall user experience.

Once again, fantastic work on the project, and I hope you find these suggestions helpful!

by @kvk-code

Allow exporting the execution results.

Options

• Text file
• PDF
• Copy to Clipboard

literallyhim — Today at 10:33 AM
chat makes 2 inputs for me, i enter 1 and second one gets stuck...anyone having same issue?
miiura — Today at 12:11 PM
Oh, that might occur depending on the situation. Are you talking about when 2 inputs are in a waiting state at the same time?
literallyhim — Today at 3:27 PM
Yes sir

from Discord

Display message if API key is not set

Improving the Waiting Time Experience

i have error in web

Unhandled Runtime Error

AxiosError: Request failed with status code 401
Call Stack
AxiosError
node_modules/axios/lib/core/AxiosError.js (22:0)
createError
node_modules/langchain/dist/util/axios-fetch-adapter.js (320:0)
settle
node_modules/langchain/dist/util/axios-fetch-adapter.js (25:0)
fetchAdapter/<
node_modules/langchain/dist/util/axios-fetch-adapter.js (175:0)
fetchAdapter
node_modules/langchain/dist/util/axios-fetch-adapter.js (167:0)

I have correctly filled all apikey .

hi there, excuse my question, completely new to this but love what you did with babyagi-ui!!! thank you very much, even with minimal starting knowledge I was able to get it to run.

my question is probably very mundane but could you advise on how to make sure one is running the most current version of baby-agi-ui?

Is it possible to have an API for it? So I can host this and use babyagi as an api

Make the image available via a Dockerfile and docker-compose.yml file to setup the environment and application.

SearpApi is expensive to use, replace with Duckduckgo.

Thank you for developing this interesting software. While testing babyagi-ui in my local environment, I encountered issues during execution of /api/context etc. The current setup instructions appear to be missing information regarding Pinecone configuration. After setting up an environment like below, the errors were resolved.

Add BabyCatAGI mode

https://github.com/yoheinakajima/babyagi/blob/main/classic/BabyCatAGI.py

Document: https://twitter.com/yoheinakajima/status/1657448504112091136

A suggestion, could we have the ability to press enter to prompt instead of having to click the icon/arrow symbol!
Would make things much more natural :)

• Current Task
• Current Task List
• Task Status
• Objective

There is a problem with installing the software package：NPM 不是内部或外部命令，也不是可执行的程序或处理文件

I received the message "ERROR_CREATING_TASKS" when trying to use GPT-4. GPT-3 it's working fine.

Hello, do you plan to adapt to the unofficial API？ (which is actually a proxy for the official web version of the API) , I currently bypass Cloudflare to call the API, support email password login, API documentation here, you can build your own tests (I can also provide API service tests)

project: https://github.com/gngpp/opengpt
api doc: https://github.com/gngpp/opengpt/blob/main/doc/rest.http

Allow BabyBeeAGI to run

https://github.com/yoheinakajima/babyagi/blob/main/classic/BabyBeeAGI

TypeError: Cannot read properties of null (reading 'useContext')

This error happened while generating the page. Any console logs will be displayed in the terminal window.

Import trace for requested module:
./node_modules/next/dist/shared/lib/head.js
Warning: Invalid hook call. Hooks can only be called inside of the body of a function component. This could happen for one of the following reasons:

1. You might have mismatching versions of React and the renderer (such as React DOM)
2. You might be breaking the Rules of Hooks
3. You might have more than one copy of React in the same app
   See https://reactjs.org/link/invalid-hook-call for tips about how to debug and fix this problem.
   [TypeError: Cannot read properties of null (reading 'useContext')]
   [TypeError: Cannot read properties of null (reading 'useContext')]

Happens for a new install and first run

TODO

• Overlapping select content
• Adjusting overall margins
• Menu open by default
• Adjust project title layout

Language support is for UI only, task results will also be supported.

Allow completion to be detected even if the screen is not in focus.

Adding Local LLMs as a Model Option.

React framework: https://github.com/r2d4/react-llm

validation for type-safe environment variables.

ref: https://env.t3.gg/, https://github.com/t3-oss/t3-env

We are getting a lot of feedback that it's typed like a ChatGPT prompt.

Hi, I keep getting this error message after following all the steps, adding the info to .env (including a active SERP API key). Any ideas how to solve this?

Unhandled Runtime Error
TypeError: searchResults is not iterable

./node_modules/next-i18next/dist/esm/config/createConfig.js 02:30:33.673 A Node.js API is used (process.cwd at line: 141) which is not supported in the Edge Runtime. 02:30:33.673 Learn more: https://nextjs.org/docs/api-reference/edge-runtime

Reference: #38 (comment)

Screenshot

We got the Translator module (#47) added, which supports the languages of the most visited countries.

• Japan
• France
• Russian
• France
• Germany
• India
• Brazil
• Thailand
• Israel
• Spain