AutoGPT icon indicating copy to clipboard operation
AutoGPT copied to clipboard

Published 20 hours ago • verified publisher icon Significant-Gravitas

AutoGPT create nothing, auto_gpt_workspace folder is empty

Open Cyrillus5 opened this issue 1 year ago • 6 comments

⚠️ Search for existing issues first ⚠️

  • [X] I have searched the existing issues, and there is no existing issue for my problem

Which Operating System are you using?

Windows

Which version of Auto-GPT are you using?

Stable (branch)

GPT-3 or GPT-4?

GPT-4

Steps to reproduce 🕹

This it is what I have inside .env :

EXECUTE_LOCAL_COMMANDS - Allow local command execution (Default: False)

RESTRICT_TO_WORKSPACE - Restrict file operations to workspace ./auto_gpt_workspace (Default: True)

EXECUTE_LOCAL_COMMANDS=True RESTRICT_TO_WORKSPACE=False

USER_AGENT - Define the user-agent used by the requests library to browse website (string)

USER_AGENT="Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.97 Safari/537.36"

AI_SETTINGS_FILE - Specifies which AI Settings file to use (defaults to ai_settings.yaml)

AI_SETTINGS_FILE=ai_settings.yaml

This it is what --debug give me : Input:python -m autogpt --debug SYSTEM: Human feedback: python -m autogpt --debug Token limit: 4000 Memory Stats: (3, (3, 1536)) Token limit: 4000 Send Token Count: 2579 Tokens remaining for response: 1421 ------------ CONTEXT SENT TO AI --------------- System: The current time and date is Sat Apr 22 23:57:42 2023

System: This reminds you of these events from your past: ['Assistant Reply: {\n "thoughts": {\n "text": "I will start by analyzing the code for the reactJS app I need to create.",\n "reasoning": "Analyzing the code will give me a better understanding of what needs to be done and how to proceed.",\n "plan": "- Analyze the code\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app",\n "criticism": "I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.",\n "speak": "I will start by analyzing the code for the reactJS app I need to create."\n },\n "command": {\n "name": "analyze_code",\n "args": {\n "code": "<full_code_string>"\n }\n }\n} \nResult: Command analyze_code returned: Error: The model: gpt-4 does not exist \nHuman Feedback: GENERATE NEXT COMMAND JSON ', 'Assistant Reply: {\n "thoughts": {\n "text": "The 'execute_shell' command did not return any output. I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine.",\n "reasoning": "The 'execute_shell_popen' command will allow me to capture the output of the command and check if Node.js is installed on this machine.",\n "plan": "- Check if Node.js is installed using the 'execute_shell_popen' command\n- If Node.js is not installed, install it\n- Analyze the code for the reactJS app\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app",\n "criticism": "I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.",\n "speak": "The 'execute_shell' command did not return any output. I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine."\n },\n "command": {\n "name": "execute_shell_popen",\n "args": {\n "command_line": "node -v"\n }\n }\n} \nResult: Human feedback: python -m autogpt --debug \nHuman Feedback: python -m autogpt --debug ', 'Assistant Reply: {\n "thoughts": {\n "text": "I will use the 'execute_shell' command to check if Node.js is installed on this machine.",\n "reasoning": "Node.js is required to run a reactJS app, so I need to make sure it is installed before proceeding.",\n "plan": "- Check if Node.js is installed using the 'execute_shell' command\n- If Node.js is not installed, install it\n- Analyze the code for the reactJS app\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app",\n "criticism": "I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.",\n "speak": "I will use the 'execute_shell' command to check if Node.js is installed on this machine."\n },\n "command": {\n "name": "execute_shell",\n "args": {\n "command_line": "node -v"\n }\n }\n} \nResult: Command execute_shell returned: None \nHuman Feedback: GENERATE NEXT COMMAND JSON ']

User: Determine which next command to use, and respond using the format specified above:

Assistant: { "thoughts": { "text": "I will start by analyzing the code for the reactJS app I need to create.", "reasoning": "Analyzing the code will give me a better understanding of what needs to be done and how to proceed.", "plan": "- Analyze the code\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app", "criticism": "I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.", "speak": "I will start by analyzing the code for the reactJS app I need to create." }, "command": { "name": "analyze_code", "args": { "code": "<full_code_string>" } } }

System: Command analyze_code returned: Error: The model: gpt-4 does not exist

User: Determine which next command to use, and respond using the format specified above:

Assistant: { "thoughts": { "text": "I will use the 'execute_shell' command to check if Node.js is installed on this machine.", "reasoning": "Node.js is required to run a reactJS app, so I need to make sure it is installed before proceeding.", "plan": "- Check if Node.js is installed using the 'execute_shell' command\n- If Node.js is not installed, install it\n- Analyze the code for the reactJS app\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app", "criticism": "I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.", "speak": "I will use the 'execute_shell' command to check if Node.js is installed on this machine." }, "command": { "name": "execute_shell", "args": { "command_line": "node -v" } } }

System: Command execute_shell returned: None

User: Determine which next command to use, and respond using the format specified above:

Assistant: { "thoughts": { "text": "The 'execute_shell' command did not return any output. I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine.", "reasoning": "The 'execute_shell_popen' command will allow me to capture the output of the command and check if Node.js is installed on this machine.", "plan": "- Check if Node.js is installed using the 'execute_shell_popen' command\n- If Node.js is not installed, install it\n- Analyze the code for the reactJS app\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app", "criticism": "I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.", "speak": "The 'execute_shell' command did not return any output. I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine." }, "command": { "name": "execute_shell_popen", "args": { "command_line": "node -v" } } }

System: Human feedback: python -m autogpt --debug

User: Determine which next command to use, and respond using the format specified above:

Current behavior 😯

Nothing was created inside auto_gpt_workspace folder however autoGPT said it writed something. Auto_gpt_workspace folder stays empty

Expected behavior 🤔

AutoGPT should write something inside auto_gpt_workspace folder.

Your prompt 📝

I don't have any file ai_settings.yaml. where it supposes to be ?

Your Logs 📒

2023-04-22 23:43:13,878 INFO NEWS:   ::UPDATED:: Welcome to Auto-GPT! We'll keep you informed of the latest news and features by printing messages here.
If you don't wish to see this message, you can run Auto-GPT with the --skip-news flag
2023-04-22 23:43:14,666 INFO WARNING:  You are running on `stable-0.2.2` branch - this is not a supported branch.
2023-04-22 23:43:15,950 INFO Welcome to Auto-GPT!  run with '--help' for more information.
2023-04-22 23:43:16,284 INFO Create an AI-Assistant: Enter the name of your AI and its role below. Entering nothing will load defaults.
2023-04-22 23:43:16,778 INFO Name your AI:  For example, 'Entrepreneur-GPT'
2023-04-22 23:43:22,658 INFO Robot here! I am at your service.
2023-04-22 23:43:22,913 INFO Describe your AI's role:  For example, 'an AI designed to autonomously develop and run businesses with the sole goal of increasing your net worth.'
2023-04-22 23:43:33,305 INFO Enter up to 5 goals for your AI:  For example: 
Increase net worth, Grow Twitter Account, Develop and manage multiple businesses autonomously'
2023-04-22 23:43:54,174 INFO Using memory of type: LocalCache
2023-04-22 23:43:54,387 INFO Using Browser: chrome
2023-04-22 23:44:08,324 INFO  THOUGHTS: I will start by analyzing the code for the reactJS app I need to create.
2023-04-22 23:44:08,797 INFO REASONING: Analyzing the code will give me a better understanding of what needs to be done and how to proceed.
2023-04-22 23:44:09,356 INFO PLAN: 
2023-04-22 23:44:09,429 INFO -  Analyze the code
2023-04-22 23:44:09,567 INFO -  Create a plan for the reactJS app
2023-04-22 23:44:09,847 INFO -  Write the code
2023-04-22 23:44:10,045 INFO -  Test the app
2023-04-22 23:44:10,210 INFO -  Deploy the app
2023-04-22 23:44:10,414 INFO CRITICISM: I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.
2023-04-22 23:44:10,963 INFO NEXT ACTION:  COMMAND = analyze_code  ARGUMENTS = {'code': '<full_code_string>'}
2023-04-22 23:44:14,757 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:44:15,554 INFO SYSTEM:  Command analyze_code returned: Error: The model: `gpt-4` does not exist
2023-04-22 23:44:27,394 INFO  THOUGHTS: I will use the 'execute_shell' command to check if Node.js is installed on this machine.
2023-04-22 23:44:27,896 INFO REASONING: Node.js is required to run a reactJS app, so I need to make sure it is installed before proceeding.
2023-04-22 23:44:28,419 INFO PLAN: 
2023-04-22 23:44:28,509 INFO -  Check if Node.js is installed
2023-04-22 23:44:28,748 INFO -  Install Node.js if it is not installed
2023-04-22 23:44:29,035 INFO -  Proceed with analyzing the code for the reactJS app
2023-04-22 23:44:29,363 INFO CRITICISM: I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.
2023-04-22 23:44:30,020 INFO NEXT ACTION:  COMMAND = execute_shell  ARGUMENTS = {'command_line': 'node -v'}
2023-04-22 23:44:42,252 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:44:43,073 INFO SYSTEM:  Command execute_shell returned: None
2023-04-22 23:44:54,173 INFO  THOUGHTS: The 'execute_shell' command did not return any output. I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine.
2023-04-22 23:44:54,775 INFO REASONING: The 'execute_shell_popen' command will allow me to capture the output of the command and check if Node.js is installed on this machine.
2023-04-22 23:44:55,334 INFO PLAN: 
2023-04-22 23:44:55,376 INFO -  Use the 'execute_shell_popen' command to check if Node.js is installed
2023-04-22 23:44:55,800 INFO -  Install Node.js if it is not installed
2023-04-22 23:44:56,112 INFO -  Proceed with analyzing the code for the reactJS app
2023-04-22 23:44:56,412 INFO CRITICISM: I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.
2023-04-22 23:44:57,133 INFO NEXT ACTION:  COMMAND = execute_shell_popen  ARGUMENTS = {'command_line': 'node -v'}
2023-04-22 23:45:14,497 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:45:15,145 INFO SYSTEM:  Command execute_shell_popen returned: Subprocess started with PID:'10828'
2023-04-22 23:45:28,189 INFO  THOUGHTS: The output of the 'execute_shell_popen' command indicates that Node.js is not installed on this machine. I will use the 'execute_shell' command to install Node.js.
2023-04-22 23:45:28,814 INFO REASONING: Node.js is required to run a reactJS app, so I need to install it before proceeding.
2023-04-22 23:45:29,291 INFO PLAN: 
2023-04-22 23:45:29,387 INFO -  Use the 'execute_shell' command to install Node.js
2023-04-22 23:45:29,714 INFO -  Proceed with analyzing the code for the reactJS app
2023-04-22 23:45:30,096 INFO CRITICISM: I need to make sure I am thorough in installing Node.js to avoid any issues when running the reactJS app.
2023-04-22 23:45:30,660 INFO NEXT ACTION:  COMMAND = execute_shell  ARGUMENTS = {'command_line': 'sudo apt-get install nodejs'}
2023-04-22 23:46:08,373 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:46:08,968 INFO SYSTEM:  Command execute_shell returned: None
2023-04-22 23:46:21,215 INFO  THOUGHTS: I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine.
2023-04-22 23:46:21,805 INFO REASONING: The 'execute_shell_popen' command will allow me to capture the output of the command and check if Node.js is installed on this machine.
2023-04-22 23:46:22,455 INFO PLAN: 
2023-04-22 23:46:22,552 INFO -  Use the 'execute_shell_popen' command to check if Node.js is installed
2023-04-22 23:46:22,886 INFO -  Proceed with analyzing the code for the reactJS app
2023-04-22 23:46:23,160 INFO CRITICISM: I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.
2023-04-22 23:46:23,750 INFO NEXT ACTION:  COMMAND = execute_shell_popen  ARGUMENTS = {'command_line': 'node -v'}
2023-04-22 23:46:41,116 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:46:41,833 INFO SYSTEM:  Command execute_shell_popen returned: Subprocess started with PID:'2768'
2023-04-22 23:46:55,391 INFO  THOUGHTS: The 'execute_shell_popen' command did not return any output. I will use the 'execute_shell_popen' command to check if Node.js is installed on this machine.
2023-04-22 23:46:56,002 INFO REASONING: The 'execute_shell_popen' command will allow me to capture the output of the command and check if Node.js is installed on this machine.
2023-04-22 23:46:56,576 INFO PLAN: 
2023-04-22 23:46:56,665 INFO -  Use the 'execute_shell_popen' command to check if Node.js is installed
2023-04-22 23:46:57,138 INFO -  Install Node.js if it is not installed
2023-04-22 23:46:57,422 INFO -  Proceed with analyzing the code for the reactJS app
2023-04-22 23:46:57,739 INFO CRITICISM: I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.
2023-04-22 23:46:58,421 INFO NEXT ACTION:  COMMAND = execute_shell_popen  ARGUMENTS = {'command_line': 'node -v'}
2023-04-22 23:53:04,525 INFO Debug Mode:  ENABLED
2023-04-22 23:53:04,763 INFO NEWS:  Welcome to Auto-GPT! We'll keep you informed of the latest news and features by printing messages here.
If you don't wish to see this message, you can run Auto-GPT with the --skip-news flag
2023-04-22 23:53:05,560 INFO WARNING:  You are running on `stable-0.2.2` branch - this is not a supported branch.
2023-04-22 23:53:06,454 INFO Welcome back!  Would you like me to return to being Robot?
2023-04-22 23:53:14,060 INFO Using memory of type: LocalCache
2023-04-22 23:53:14,267 INFO Using Browser: chrome
2023-04-22 23:53:14,407 INFO Prompt: You are Robot, Create a reactJS app
Your decisions must always be made independently without seeking user assistance. Play to your strengths as an LLM and pursue simple strategies with no legal complications.

GOALS:

1. create a simple reactjs app
2. nothing
3. nothing
4. nothing
5. nothing


Constraints:
1. ~4000 word limit for short term memory. Your short term memory is short, so immediately save important information to files.
2. If you are unsure how you previously did something or want to recall past events, thinking about similar events will help you remember.
3. No user assistance
4. Exclusively use the commands listed in double quotes e.g. "command name"

Commands:
1. analyze_code: Analyze Code, args: "code": "<full_code_string>"
2. execute_python_file: Execute Python File, args: "filename": "<filename>"
3. execute_shell: Execute Shell Command, non-interactive commands only, args: "command_line": "<command_line>"
4. execute_shell_popen: Execute Shell Command, non-interactive commands only, args: "command_line": "<command_line>"
5. append_to_file: Append to file, args: "filename": "<filename>", "text": "<text>"
6. delete_file: Delete file, args: "filename": "<filename>"
7. read_file: Read file, args: "filename": "<filename>"
8. search_files: Search Files, args: "directory": "<directory>"
9. write_to_file: Write to file, args: "filename": "<filename>", "text": "<text>"
10. google: Google Search, args: "query": "<query>"
11. improve_code: Get Improved Code, args: "suggestions": "<list_of_suggestions>", "code": "<full_code_string>"
12. send_tweet: Send Tweet, args: "tweet_text": "<tweet_text>"
13. browse_website: Browse Website, args: "url": "<url>", "question": "<what_you_want_to_find_on_website>"
14. write_tests: Write Tests, args: "code": "<full_code_string>", "focus": "<list_of_focus_areas>"
15. delete_agent: Delete GPT Agent, args: "key": "<key>"
16. get_hyperlinks: Get text summary, args: "url": "<url>"
17. get_text_summary: Get text summary, args: "url": "<url>", "question": "<question>"
18. list_agents: List GPT Agents, args: () -> str
19. message_agent: Message GPT Agent, args: "key": "<key>", "message": "<message>"
20. start_agent: Start GPT Agent, args: "name": "<name>", "task": "<short_task_desc>", "prompt": "<prompt>"
21. Do Nothing: "do_nothing", args: 
22. Task Complete (Shutdown): "task_complete", args: "reason": "<reason>"

Resources:
1. Internet access for searches and information gathering.
2. Long Term memory management.
3. GPT-3.5 powered Agents for delegation of simple tasks.
4. File output.

Performance Evaluation:
1. Continuously review and analyze your actions to ensure you are performing to the best of your abilities.
2. Constructively self-criticize your big-picture behavior constantly.
3. Reflect on past decisions and strategies to refine your approach.
4. Every command has a cost, so be smart and efficient. Aim to complete tasks in the least number of steps.
5. Write all code to a file.

You should only respond in JSON format as described below 
Response Format: 
{
    "thoughts": {
        "text": "thought",
        "reasoning": "reasoning",
        "plan": "- short bulleted\n- list that conveys\n- long-term plan",
        "criticism": "constructive self-criticism",
        "speak": "thoughts summary to say to user"
    },
    "command": {
        "name": "command name",
        "args": {
            "arg name": "value"
        }
    }
} 
Ensure the response can beparsed by Python json.loads
2023-04-22 23:53:20,481 DEBUG  Token limit: 4000
2023-04-22 23:53:20,481 DEBUG  Memory Stats: (0, (0, 1536))
2023-04-22 23:53:20,980 DEBUG  Token limit: 4000
2023-04-22 23:53:20,980 DEBUG  Send Token Count: 943
2023-04-22 23:53:20,980 DEBUG  Tokens remaining for response: 3057
2023-04-22 23:53:20,985 DEBUG  ------------ CONTEXT SENT TO AI ---------------
2023-04-22 23:53:20,985 DEBUG  System: The current time and date is Sat Apr 22 23:53:20 2023
2023-04-22 23:53:20,985 DEBUG  
2023-04-22 23:53:20,987 DEBUG  System: This reminds you of these events from your past:



2023-04-22 23:53:20,988 DEBUG  
2023-04-22 23:53:20,988 DEBUG  User: Determine which next command to use, and respond using the format specified above:
2023-04-22 23:53:20,988 DEBUG  
2023-04-22 23:53:20,989 DEBUG  ----------- END OF CONTEXT ----------------
2023-04-22 23:53:31,988 INFO  THOUGHTS: I will start by analyzing the code for the reactJS app I need to create.
2023-04-22 23:53:32,506 INFO REASONING: Analyzing the code will give me a better understanding of what needs to be done and how to proceed.
2023-04-22 23:53:33,030 INFO PLAN: 
2023-04-22 23:53:33,087 INFO -  Analyze the code
2023-04-22 23:53:33,249 INFO -  Create a plan for the reactJS app
2023-04-22 23:53:33,568 INFO -  Write the code
2023-04-22 23:53:33,739 INFO -  Test the app
2023-04-22 23:53:33,884 INFO -  Deploy the app
2023-04-22 23:53:34,095 INFO CRITICISM: I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.
2023-04-22 23:53:34,660 INFO NEXT ACTION:  COMMAND = analyze_code  ARGUMENTS = {'code': '<full_code_string>'}
2023-04-22 23:53:49,102 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:53:49,890 INFO SYSTEM:  Command analyze_code returned: Error: The model: `gpt-4` does not exist
2023-04-22 23:53:50,218 DEBUG  Token limit: 4000
2023-04-22 23:53:50,504 DEBUG  Memory Stats: (1, (1, 1536))
2023-04-22 23:53:50,516 DEBUG  Token limit: 4000
2023-04-22 23:53:50,516 DEBUG  Send Token Count: 1416
2023-04-22 23:53:50,516 DEBUG  Tokens remaining for response: 2584
2023-04-22 23:53:50,516 DEBUG  ------------ CONTEXT SENT TO AI ---------------
2023-04-22 23:53:50,516 DEBUG  System: The current time and date is Sat Apr 22 23:53:50 2023
2023-04-22 23:53:50,520 DEBUG  
2023-04-22 23:53:50,520 DEBUG  System: This reminds you of these events from your past:
['Assistant Reply: {\n    "thoughts": {\n        "text": "I will start by analyzing the code for the reactJS app I need to create.",\n        "reasoning": "Analyzing the code will give me a better understanding of what needs to be done and how to proceed.",\n        "plan": "- Analyze the code\\n- Create a plan for the reactJS app\\n- Write the code\\n- Test the app\\n- Deploy the app",\n        "criticism": "I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.",\n        "speak": "I will start by analyzing the code for the reactJS app I need to create."\n    },\n    "command": {\n        "name": "analyze_code",\n        "args": {\n            "code": "<full_code_string>"\n        }\n    }\n} \nResult: Command analyze_code returned: Error: The model: `gpt-4` does not exist \nHuman Feedback: GENERATE NEXT COMMAND JSON ']


2023-04-22 23:53:50,524 DEBUG  
2023-04-22 23:53:50,525 DEBUG  User: Determine which next command to use, and respond using the format specified above:
2023-04-22 23:53:50,525 DEBUG  
2023-04-22 23:53:50,525 DEBUG  Assistant: {
    "thoughts": {
        "text": "I will start by analyzing the code for the reactJS app I need to create.",
        "reasoning": "Analyzing the code will give me a better understanding of what needs to be done and how to proceed.",
        "plan": "- Analyze the code\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app",
        "criticism": "I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.",
        "speak": "I will start by analyzing the code for the reactJS app I need to create."
    },
    "command": {
        "name": "analyze_code",
        "args": {
            "code": "<full_code_string>"
        }
    }
}
2023-04-22 23:53:50,533 DEBUG  
2023-04-22 23:53:50,533 DEBUG  System: Command analyze_code returned: Error: The model: `gpt-4` does not exist
2023-04-22 23:53:50,535 DEBUG  
2023-04-22 23:53:50,536 DEBUG  User: Determine which next command to use, and respond using the format specified above:
2023-04-22 23:53:50,537 DEBUG  
2023-04-22 23:53:50,538 DEBUG  ----------- END OF CONTEXT ----------------
2023-04-22 23:54:03,203 INFO  THOUGHTS: I will use the 'execute_shell' command to check if Node.js is installed on this machine.
2023-04-22 23:54:03,743 INFO REASONING: Node.js is required to run a reactJS app, so I need to make sure it is installed before proceeding.
2023-04-22 23:54:04,288 INFO PLAN: 
2023-04-22 23:54:04,382 INFO -  Check if Node.js is installed using the 'execute_shell' command
2023-04-22 23:54:04,715 INFO -  If Node.js is not installed, install it
2023-04-22 23:54:04,985 INFO -  Analyze the code for the reactJS app
2023-04-22 23:54:05,336 INFO -  Create a plan for the reactJS app
2023-04-22 23:54:05,640 INFO -  Write the code
2023-04-22 23:54:05,819 INFO -  Test the app
2023-04-22 23:54:06,030 INFO -  Deploy the app
2023-04-22 23:54:06,220 INFO CRITICISM: I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.
2023-04-22 23:54:06,807 INFO NEXT ACTION:  COMMAND = execute_shell  ARGUMENTS = {'command_line': 'node -v'}
2023-04-22 23:56:23,853 INFO -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= 
2023-04-22 23:56:24,481 INFO SYSTEM:  Command execute_shell returned: None
2023-04-22 23:56:24,673 DEBUG  Token limit: 4000
2023-04-22 23:56:25,093 DEBUG  Memory Stats: (2, (2, 1536))
2023-04-22 23:56:25,104 DEBUG  Token limit: 4000
2023-04-22 23:56:25,104 DEBUG  Send Token Count: 1953
2023-04-22 23:56:25,107 DEBUG  Tokens remaining for response: 2047
2023-04-22 23:56:25,110 DEBUG  ------------ CONTEXT SENT TO AI ---------------
2023-04-22 23:56:25,110 DEBUG  System: The current time and date is Sat Apr 22 23:56:25 2023
2023-04-22 23:56:25,110 DEBUG  
2023-04-22 23:56:25,110 DEBUG  System: This reminds you of these events from your past:
['Assistant Reply: {\n    "thoughts": {\n        "text": "I will start by analyzing the code for the reactJS app I need to create.",\n        "reasoning": "Analyzing the code will give me a better understanding of what needs to be done and how to proceed.",\n        "plan": "- Analyze the code\\n- Create a plan for the reactJS app\\n- Write the code\\n- Test the app\\n- Deploy the app",\n        "criticism": "I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.",\n        "speak": "I will start by analyzing the code for the reactJS app I need to create."\n    },\n    "command": {\n        "name": "analyze_code",\n        "args": {\n            "code": "<full_code_string>"\n        }\n    }\n} \nResult: Command analyze_code returned: Error: The model: `gpt-4` does not exist \nHuman Feedback: GENERATE NEXT COMMAND JSON ', 'Assistant Reply: {\n    "thoughts": {\n        "text": "I will use the \'execute_shell\' command to check if Node.js is installed on this machine.",\n        "reasoning": "Node.js is required to run a reactJS app, so I need to make sure it is installed before proceeding.",\n        "plan": "- Check if Node.js is installed using the \'execute_shell\' command\\n- If Node.js is not installed, install it\\n- Analyze the code for the reactJS app\\n- Create a plan for the reactJS app\\n- Write the code\\n- Test the app\\n- Deploy the app",\n        "criticism": "I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.",\n        "speak": "I will use the \'execute_shell\' command to check if Node.js is installed on this machine."\n    },\n    "command": {\n        "name": "execute_shell",\n        "args": {\n            "command_line": "node -v"\n        }\n    }\n} \nResult: Command execute_shell returned: None \nHuman Feedback: GENERATE NEXT COMMAND JSON ']


2023-04-22 23:56:25,117 DEBUG  
2023-04-22 23:56:25,118 DEBUG  User: Determine which next command to use, and respond using the format specified above:
2023-04-22 23:56:25,118 DEBUG  
2023-04-22 23:56:25,118 DEBUG  Assistant: {
    "thoughts": {
        "text": "I will start by analyzing the code for the reactJS app I need to create.",
        "reasoning": "Analyzing the code will give me a better understanding of what needs to be done and how to proceed.",
        "plan": "- Analyze the code\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app",
        "criticism": "I need to make sure I am thorough in my analysis and planning to avoid any mistakes in the code.",
        "speak": "I will start by analyzing the code for the reactJS app I need to create."
    },
    "command": {
        "name": "analyze_code",
        "args": {
            "code": "<full_code_string>"
        }
    }
}
2023-04-22 23:56:25,118 DEBUG  
2023-04-22 23:56:25,118 DEBUG  System: Command analyze_code returned: Error: The model: `gpt-4` does not exist
2023-04-22 23:56:25,118 DEBUG  
2023-04-22 23:56:25,125 DEBUG  User: Determine which next command to use, and respond using the format specified above:
2023-04-22 23:56:25,127 DEBUG  
2023-04-22 23:56:25,127 DEBUG  Assistant: {
    "thoughts": {
        "text": "I will use the 'execute_shell' command to check if Node.js is installed on this machine.",
        "reasoning": "Node.js is required to run a reactJS app, so I need to make sure it is installed before proceeding.",
        "plan": "- Check if Node.js is installed using the 'execute_shell' command\n- If Node.js is not installed, install it\n- Analyze the code for the reactJS app\n- Create a plan for the reactJS app\n- Write the code\n- Test the app\n- Deploy the app",
        "criticism": "I need to make sure I am thorough in checking if Node.js is installed to avoid any issues when running the reactJS app.",
        "speak": "I will use the 'execute_shell' command to check if Node.js is installed on this machine."
    },
    "command": {
        "name": "execute_shell",
        "args": {
            "command_line": "node -v"
        }
    }
}
2023-04-22 23:56:25,132 DEBUG  
2023-04-22 23:56:25,132 DEBUG  System: Command execute_shell returned: None
2023-04-22 23:56:25,135 DEBUG  
2023-04-22 23:56:25,135 DEBUG  User: Determine which next command to use, and respond using the format specified above:
2023-04-22 23:56:25,135 DEBUG  
2023-04-22 23:56:25,136 DEBUG  ----------- END OF CONTEXT ----------------

Cyrillus5 avatar Apr 23 '23 16:04 Cyrillus5

Any suggestions ?

Cyrillus5 avatar Apr 24 '23 09:04 Cyrillus5

I have the same issue. I have no workspace folder. No JSON file and anytime auto GPT says it created a file and completed the task, there is nothing to be found.

Nexenn avatar May 01 '23 15:05 Nexenn

Me 2, don't know why.

I created a file .env : EXECUTE_LOCAL_COMMANDS=True OPENAI_API_KEY=my-api-key-aaa EXECUTE_LOCAL_COMMANDS=True

Then when I launch the docker : docker run -it --env-file=.env --rm auto-gpt_auto-gpt --gpt3only --continuous

it said : Warning: The file 'auto-gpt.json' does not exist. Local memory would not be saved to a file. Continuous Mode: ENABLED WARNING: Continuous mode is not recommended. It is potentially dangerous and may cause your AI to run forever or carry out actions you would not usually authorise. Use at your own risk. GPT3.5 Only Mode: ENABLED

When I ask autogpt to code and save the results in files, he does nothing.

aDubanchet avatar May 03 '23 10:05 aDubanchet

As part of #3911 it would be possible to provide a "setup mode" to run a wizard to check some heuristics to see if the workspace/autogpt.json stuff is there or not, and if necessary re-generate missing stuff.

The idea would be to come up with a dedicated workflow for fixing some of the more common issues. Feedback welcome!

Boostrix avatar May 06 '23 20:05 Boostrix

also having this issue

andreiashu avatar May 08 '23 06:05 andreiashu

Shame on me, i thought that I pulled the stable version Test if you are on the stable version : git branch

aDubanchet avatar May 08 '23 10:05 aDubanchet

Writing results to file did not work (as file contained only python code) although AutoGPT reported "Task completed successfully" and was shutting down.

Nothing was printed to console, although such a task existed (below)

Prompt: SPEAK: I will create a Python script to read the file containing the names, addresses and services of the primary care private clinics and print the information to the console. NEXT ACTION: COMMAND = write_to_file ARGUMENTS = {'filename': '/app/autogpt/auto_gpt_workspace/print_info.py', 'text': "with open('clinic_info_filtered.txt', 'r') as f:\n data = f.read()\n print(data)"} Enter 'y' to authorise command, 'y -N' to run N continuous commands, 's' to run self-feedback commands, 'n' to exit program, or enter feedback for TestGPT... Asking user via keyboard... Input:y -5 -=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-= SYSTEM: Command write_to_file returned: File written to successfully. TESTGPT THOUGHTS: I will use the task_complete command to end this session. REASONING: Using the task_complete command will end this session and free up resources for other tasks. PLAN:

  • Use the task_complete command to end this session CRITICISM: None SPEAK: I will use the task_complete command to end this session. NEXT ACTION: COMMAND = task_complete ARGUMENTS = {'reason': 'Task completed successfully.'} AUTHORISED COMMANDS LEFT: 4 Shutting down... Task completed successfully.

And the results file "CLINIC_INFO_FILTERED.TXT" contains only:

"import re

Read file containing clinic information with open('clinic_info.txt', 'r') as f: data = f.read()

Exclude private hospitals filtered_data = re.sub(r'Private Hospital', '', data)

Save filtered information to file with open('clinic_info_filtered.txt', 'w') as f: f.write(filtered_data) "

file "CLINIC_INFO.TXT" does not exist

KittHaab avatar Jun 17 '23 15:06 KittHaab

So I'm having the same issue, where AutoGPT says it is writing to files, but the auto_gpt_workspace folder (which is located where my docker-compose.yml is located) stays empty no matter what.

I then realized after looking into the docker container while the project is running, autogpt is in fact writing files to this directory /app/autogpt/workspace/auto_gpt_workspace . Though it's only accessible via the running docker container via Terminal. Though due to the nature of docker containers, as soon as you exit the running AutoGPT, you will lose any documents it creates. So it could be that running this project via docker has a particular issue moving the files back out whenever it completes a write to a file.

I'm totally new to AutoGPT, I just set it up yesterday & I will try to investigate why this issue is happening.

bchehraz avatar Jul 10 '23 20:07 bchehraz

https://github.com/Significant-Gravitas/Auto-GPT/pull/5013 this should fix it, the correct docker-compose.yml in the docs should contain the extra "workspace" folder in the path:

version: "3.9"
services:
 auto-gpt:
   image: significantgravitas/auto-gpt
   env_file:
     - .env
   profiles: ["exclude-from-up"]
   volumes:
     - ./auto_gpt_workspace:/app/autogpt/workspace/auto_gpt_workspace
     - ./data:/app/data
     ## allow auto-gpt to write logs to disk
     - ./logs:/app/logs
     ## uncomment following lines if you want to make use of these files
     ## you must have them existing in the same folder as this docker-compose.yml
     #- type: bind
     #  source: ./azure.yaml
     #  target: /app/azure.yaml
     #- type: bind
     #  source: ./ai_settings.yaml
     #  target: /app/ai_settings.yaml

0xunreal avatar Jul 19 '23 13:07 0xunreal

Fixed in #5032 :)

Pwuts avatar Jul 21 '23 18:07 Pwuts