gpt-engineer
gpt-engineer copied to clipboard
I answer a question, then it says ''No'' and proceeds without clarifying anything else + Hallucinating
This keeps happening...
PS C:\Users\jews4\OneDrive\one2\OneDrive\Documents\gpt-engineer-main> python main.py gpt-zdondada-builds Areas that need clarification:
- Login/Register screen design and layout
- Main Search For Prescribers screen design and layout
- Account/Credits screen design and layout
- Contact Us screen design and layout
- App Bar design and functionality
- Field mapping for search results and More Info modal
- User registration process and required information
- Credits purchase process and options
Clarifying question: For the main Search For Prescribers screen, how should the search results be displayed? Should they be in a table format or a card layout? (answer in text, or "q" to move on) table
No Core classes, functions, and methods:
- App.vue - Main Vue application component
- router/index.js - Vue Router configuration
- store/index.js - Vuex store configuration
- components/LoginRegister.vue - Login/Register component
Aborted! 5. components/SearchPrescribers
ALSO, if I let it run through the project, it outputs a couple of things and then tells me the rest of the files are in a non-existent GitHub
This was in the logs if it helps:
{"role": "assistant", "content": "Areas that need clarification:\n1. Login/Register screen design and layout\n2. Main Search For Prescribers screen design and layout\n3. Account/Credits screen design and layout\n4. Contact Us screen design and layout\n5. App Bar design and functionality\n6. Field mapping for search results and More Info modal\n7. User registration process and required information\n8. Credits purchase process and options\n\nClarifying question:\nFor the main Search For Prescribers screen, how should the search results be displayed? Should they be in a table format or a card layout?"}, {"role": "user", "content": "table\n\nIs anything else unclear? If yes, only answer in the form:\n{remaining unclear areas} remaining questions.\n{Next question}\nIf everything is sufficiently clear, only answer "no"."}, {"role": "assistant", "content": "No"}]
I'm getting this too. It might be a token limiter?
Whenever and I paste it gives a NO
I was giving typed replies and got four questions in before getting a NO
Tried again, same thing, here's another example of the hallucinating
Please find the code for the components and API functions in the following files:
I think this concept needs some tiered agent creation to solve intermediate tasks
Fixed in https://github.com/AntonOsika/gpt-engineer/pull/21
I'm also getting this issue, where it simply replies "no" and then it packs up it's stuff and goes home (it just terminates with no errors and ceases to do anymore development/building...sometimes it gets through only one of its clarification questions, sometimes multiple...but it always seems to happen before it gets answers to them all.
still getting a "no" after clarifying it the task and just runs without reclarification of other tasks
I think this was just fixed in 6a3f7eb