Multiple file code improvements. Need a Backup feature...

[ChessScholar]

I changed the coding in a few files for stability and error checks. just change all of the extensions from .txt to .py except the prompt.

I did my best to keep the coding universal.

I highly recommend completing the other call functions as described in Homepage Auto-GPT . such as setting up all the APIs. They are pretty easy.

This fixed a lot of the JSON errors, parsing issues, and Unknown command Errors. There are still a lot of things that need to be improved.

json_parser.py (this was very annoying to debug)

• Added import re
• Added def extract_json for checks on the error of "none" value.
• Removed some of the # fluff.
• Fixed the print error for json error checks.

prompt.txt (I don't know if there was anything different, but my "date created" showed the same as the other files, so I included it lol.

main.txt

• Fixed minor typing speed (min/max), where min was higher than max.
• raise an exception if pinecone_api_key or region is not provided if not cfg.pinecone_api_key or not cfg.pinecone_region: raise Exception ("Please provide pinecone_api_key and pinecone_region") has not yet been added after cfg = Config()
• Line 374, added memory_to_add = f"Assistant Reply: {assistant_reply} " \ f"\nResult: {result} " \ f"\nHuman Feedback: {user_input} "

commands.txt

• Added from call_ai_function import call_ai_function
• Added model_name = 'GPT-4' for "cannot find gpt4" issues. (fake solution :P)
• Line 104 added elif command_name == "think": return think(arguments["topic"])
• Line 283, added def think(topic): [long code]

json_parser Improved.txt
prompt.txt
main.txt
commands.txt

To be desired: A backup function for information already learned to call back on if the system crashes (usually from a Gateway error).

[deronkel82]

Hi! Thanks for the changes. Sadly i still get the error:

Error:
Traceback (most recent call last): File "D:\KI\Auto-GPT-master\scripts\main.py", line 67, in print_assistant_thoughts assistant_thoughts = assistant_reply_json.get("thoughts", {}) AttributeError: 'NoneType' object has no attribute 'get'
Error parsing JSON: Expecting ',' delimiter: line 12 column 153 (char 1692)
NEXT ACTION: COMMAND = Error: ARGUMENTS = argument of type 'NoneType' is not iterable

Any ideas?

[InventorCoin]

Input:SYSTEM: Human feedback:
No JSON found in the input string.
Error:
Traceback (most recent call last): File "/Users/markausten/Documents/GitHub/autogtp4/Auto-GPT-master/scripts/main.py", line 67, in print_assistant_thoughts assistant_thoughts = assistant_reply_json.get("thoughts", {}) ^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'NoneType' object has no attribute 'get'
No JSON found in the input string.
NEXT ACTION: COMMAND = Error: ARGUMENTS = argument of type 'NoneType' is not iterable
Enter 'y' to authorise command, 'y -N' to run N continuous commands, 'n' to exit program, or enter feedback for test...
Input:

[lupuleasa-core]

@ChessScholar
chess, can you not create pull requests instead of discussions?

install github desktop https://desktop.github.com/

and press Ctrl+R inside to create pull request

[ChessScholar]

Thanks! Brand new to using Github, not familiar with all this.