diff --git a/README.md b/README.md index 5848c585..d66a6022 100644 --- a/README.md +++ b/README.md @@ -20,10 +20,11 @@ Your support is greatly appreciated Development of this free, open-source project is made possible by all the contributors and sponsors. If you'd like to sponsor this project and have your avatar or company logo appear below click here. 💖

-nocodeclarity  tjarmain  tekelsey  robinicus  digisomni   +thepok  SpacingLily  m  zkonduit  maxxflyer  tekelsey  nocodeclarity  tjarmain  alexisneuhaus  jaumebalust  robinicus  digisomni  

+

-alexisneuhaus  iokode  jaumebalust   +alexisneuhaus  iokode  jaumebalust  nova-land  robinicus  Void-n-Null  ritesh24  merwanehamadi  raulmarindev  siduppal  goosecubedaddy  pleabargain  

diff --git a/requirements-alternative.txt b/requirements-alternative.txt deleted file mode 100644 index a2f2723d..00000000 --- a/requirements-alternative.txt +++ /dev/null @@ -1,14 +0,0 @@ -# I wasn't having any luck installing the requirements.txt file in Mac or Linux -# But this seems to work. -# The biggest difference is docker 5 instead of 6, because of this silliness: -# -# The conflict is caused by: -# The user requested requests>=2.26.0 -# docker 6.0.1 depends on requests>=2.26.0 -# googlesearch-python 1.1.0 depends on requests==2.25.1 -docker==5.0.3 - -# I'd love to fix this in a cleaner way - -# Now go ahead and install the rest of what requirements.txt says: --r requirements.txt diff --git a/requirements.txt b/requirements.txt index 063931a9..158e9324 100644 --- a/requirements.txt +++ b/requirements.txt @@ -7,7 +7,7 @@ pyyaml==6.0 readability-lxml==0.8.1 requests tiktoken==0.3.3 +gTTS==2.3.1 docker -googlesearch-python +duckduckgo-search google-api-python-client #(https://developers.google.com/custom-search/v1/overview) -# Googlesearch python seems to be a bit cursed, anyone good at fixing thigns like this? \ No newline at end of file diff --git a/scripts/browse.py b/scripts/browse.py index f096c5f3..510f9c29 100644 --- a/scripts/browse.py +++ b/scripts/browse.py @@ -74,30 +74,25 @@ def split_text(text, max_length=8192): yield "\n".join(current_chunk) -def summarize_text(text, is_website=True): - if text == "": +def create_message(chunk, question): + return { + "role": "user", + "content": f"\"\"\"{chunk}\"\"\" Using the above text, please answer the following question: \"{question}\" -- if the question cannot be answered using the text, please summarize the text." + } + +def summarize_text(text, question): + if not text: return "Error: No text to summarize" - print("Text length: " + str(len(text)) + " characters") + text_length = len(text) + print(f"Text length: {text_length} characters") + summaries = [] chunks = list(split_text(text)) for i, chunk in enumerate(chunks): - print("Summarizing chunk " + str(i + 1) + " / " + str(len(chunks))) - if is_website: - messages = [ - { - "role": "user", - "content": "Please summarize the following website text, do not describe the general website, but instead concisely extract the specific information this subpage contains.: " + - chunk}, - ] - else: - messages = [ - { - "role": "user", - "content": "Please summarize the following text, focusing on extracting concise and specific information: " + - chunk}, - ] + print(f"Summarizing chunk {i + 1} / {len(chunks)}") + messages = [create_message(chunk, question)] summary = create_chat_completion( model=cfg.fast_llm_model, @@ -105,25 +100,11 @@ def summarize_text(text, is_website=True): max_tokens=300, ) summaries.append(summary) - print("Summarized " + str(len(chunks)) + " chunks.") + + print(f"Summarized {len(chunks)} chunks.") combined_summary = "\n".join(summaries) - - # Summarize the combined summary - if is_website: - messages = [ - { - "role": "user", - "content": "Please summarize the following website text, do not describe the general website, but instead concisely extract the specific information this subpage contains.: " + - combined_summary}, - ] - else: - messages = [ - { - "role": "user", - "content": "Please summarize the following text, focusing on extracting concise and specific infomation: " + - combined_summary}, - ] + messages = [create_message(combined_summary, question)] final_summary = create_chat_completion( model=cfg.fast_llm_model, @@ -131,4 +112,4 @@ def summarize_text(text, is_website=True): max_tokens=300, ) - return final_summary + return final_summary \ No newline at end of file diff --git a/scripts/commands.py b/scripts/commands.py index ffb88d61..6000cbdb 100644 --- a/scripts/commands.py +++ b/scripts/commands.py @@ -9,7 +9,7 @@ import ai_functions as ai from file_operations import read_file, write_to_file, append_to_file, delete_file from execute_code import execute_python_file from json_parser import fix_and_parse_json -from googlesearch import search +from duckduckgo_search import ddg from googleapiclient.discovery import build from googleapiclient.errors import HttpError @@ -72,7 +72,7 @@ def execute_command(command_name, arguments): elif command_name == "delete_agent": return delete_agent(arguments["key"]) elif command_name == "get_text_summary": - return get_text_summary(arguments["url"]) + return get_text_summary(arguments["url"], arguments["question"]) elif command_name == "get_hyperlinks": return get_hyperlinks(arguments["url"]) elif command_name == "read_file": @@ -84,7 +84,7 @@ def execute_command(command_name, arguments): elif command_name == "delete_file": return delete_file(arguments["file"]) elif command_name == "browse_website": - return browse_website(arguments["url"]) + return browse_website(arguments["url"], arguments["question"]) # TODO: Change these to take in a file rather than pasted code, if # non-file is given, return instructions "Input should be a python # filepath, write your code to file and try again" @@ -112,7 +112,7 @@ def get_datetime(): def google_search(query, num_results=8): search_results = [] - for j in search(query, num_results=num_results): + for j in ddg(query, max_results=num_results): search_results.append(j) return json.dumps(search_results, ensure_ascii=False, indent=4) @@ -152,8 +152,8 @@ def google_official_search(query, num_results=8): # Return the list of search result URLs return search_results_links -def browse_website(url): - summary = get_text_summary(url) +def browse_website(url, question): + summary = get_text_summary(url, question) links = get_hyperlinks(url) # Limit links to 5 @@ -165,9 +165,9 @@ def browse_website(url): return result -def get_text_summary(url): +def get_text_summary(url, question): text = browse.scrape_text(url) - summary = browse.summarize_text(text) + summary = browse.summarize_text(text, question) return """ "Result" : """ + summary diff --git a/scripts/config.py b/scripts/config.py index 51e11757..766cb94f 100644 --- a/scripts/config.py +++ b/scripts/config.py @@ -61,7 +61,7 @@ class Config(metaclass=Singleton): self.smart_token_limit = value def set_openai_api_key(self, value: str): - self.apiopenai_api_key_key = value + self.openai_api_key = value def set_elevenlabs_api_key(self, value: str): self.elevenlabs_api_key = value diff --git a/scripts/data.py b/scripts/data.py index 0a72cbbf..72a6bbfc 100644 --- a/scripts/data.py +++ b/scripts/data.py @@ -1,6 +1,6 @@ import os from pathlib import Path - +SRC_DIR = Path(__file__).parent def load_prompt(): try: @@ -9,7 +9,7 @@ def load_prompt(): data_dir = file_dir / "data" prompt_file = data_dir / "prompt.txt" # Load the promt from data/prompt.txt - with open(prompt_file, "r") as prompt_file: + with open(SRC_DIR/ "data/prompt.txt", "r") as prompt_file: prompt = prompt_file.read() return prompt diff --git a/scripts/data/prompt.txt b/scripts/data/prompt.txt index d17fa27a..a93e783e 100644 --- a/scripts/data/prompt.txt +++ b/scripts/data/prompt.txt @@ -1,7 +1,8 @@ CONSTRAINTS: -1. ~4000 word limit for memory. Your memory is short, so immidiately save important information to long term memory and code to files. +1. ~4000 word limit for memory. Your memory is short, so immediately save important information to long term memory and code to files. 2. No user assistance +3. Exclusively use the commands listed in double quotes e.g. "command name" COMMANDS: @@ -9,7 +10,7 @@ COMMANDS: 2. Memory Add: "memory_add", args: "string": "" 3. Memory Delete: "memory_del", args: "key": "" 4. Memory Overwrite: "memory_ovr", args: "key": "", "string": "" -5. Browse Website: "browse_website", args: "url": "" +5. Browse Website: "browse_website", args: "url": "", "question": "" 6. Start GPT Agent: "start_agent", args: "name": , "task": "", "prompt": "" 7. Message GPT Agent: "message_agent", args: "key": "", "message": "" 8. List GPT Agents: "list_agents", args: "" @@ -34,9 +35,9 @@ RESOURCES: 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 behaviour constantly. +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 efficent. Aim to complete tasks in the least number of steps. +4. Every command has a cost, so be smart and efficient. Aim to complete tasks in the least number of steps. You should only respond in JSON format as described below @@ -58,4 +59,4 @@ RESPONSE FORMAT: } } -Ensure the response can be parsed by Python json.loads \ No newline at end of file +Ensure the response can be parsed by Python json.loads diff --git a/scripts/main.py b/scripts/main.py index 97c05d7d..93124234 100644 --- a/scripts/main.py +++ b/scripts/main.py @@ -52,6 +52,14 @@ def print_assistant_thoughts(assistant_reply): # Parse and print Assistant response assistant_reply_json = fix_and_parse_json(assistant_reply) + # Check if assistant_reply_json is a string and attempt to parse it into a JSON object + if isinstance(assistant_reply_json, str): + try: + assistant_reply_json = json.loads(assistant_reply_json) + except json.JSONDecodeError as e: + print_to_console("Error: Invalid JSON\n", Fore.RED, assistant_reply) + assistant_reply_json = {} + assistant_thoughts_reasoning = None assistant_thoughts_plan = None assistant_thoughts_speak = None @@ -303,7 +311,7 @@ while True: Fore.CYAN, f"COMMAND = {Fore.CYAN}{command_name}{Style.RESET_ALL} ARGUMENTS = {Fore.CYAN}{arguments}{Style.RESET_ALL}") print( - "Enter 'y' to authorise command or 'n' to exit program...", + f"Enter 'y' to authorise command or 'n' to exit program, or enter feedback for {ai_name}...", flush=True) while True: console_input = input(Fore.MAGENTA + "Input:" + Style.RESET_ALL) @@ -314,16 +322,18 @@ while True: user_input = "EXIT" break else: - continue + user_input = console_input + command_name = "human_feedback" + break - if user_input != "GENERATE NEXT COMMAND JSON": - print("Exiting...", flush=True) - break - - print_to_console( + if user_input == "GENERATE NEXT COMMAND JSON": + print_to_console( "-=-=-=-=-=-=-= COMMAND AUTHORISED BY USER -=-=-=-=-=-=-=", Fore.MAGENTA, "") + elif user_input == "EXIT": + print("Exiting...", flush=True) + break else: # Print command print_to_console( @@ -332,10 +342,12 @@ while True: f"COMMAND = {Fore.CYAN}{command_name}{Style.RESET_ALL} ARGUMENTS = {Fore.CYAN}{arguments}{Style.RESET_ALL}") # Execute command - if command_name.lower() != "error": - result = f"Command {command_name} returned: {cmd.execute_command(command_name, arguments)}" - else: + if command_name.lower() == "error": result = f"Command {command_name} threw the following error: " + arguments + elif command_name == "human_feedback": + result = f"Human feedback: {user_input}" + else: + result = f"Command {command_name} returned: {cmd.execute_command(command_name, arguments)}" # Check if there's a result from the command append it to the message # history @@ -347,3 +359,4 @@ while True: chat.create_chat_message( "system", "Unable to execute command")) print_to_console("SYSTEM: ", Fore.YELLOW, "Unable to execute command") + diff --git a/scripts/speak.py b/scripts/speak.py index 2fbcbed2..f6242a37 100644 --- a/scripts/speak.py +++ b/scripts/speak.py @@ -3,6 +3,8 @@ from playsound import playsound import requests from config import Config cfg = Config() +import gtts + # TODO: Nicer names for these ids voices = ["ErXwobaYiN019PkySvjV", "EXAVITQu4vr4xnSDxMaL"] @@ -12,10 +14,9 @@ tts_headers = { "xi-api-key": cfg.elevenlabs_api_key } -def say_text(text, voice_index=0): +def eleven_labs_speech(text, voice_index=0): tts_url = "https://api.elevenlabs.io/v1/text-to-speech/{voice_id}".format( voice_id=voices[voice_index]) - formatted_message = {"text": text} response = requests.post( tts_url, headers=tts_headers, json=formatted_message) @@ -24,8 +25,24 @@ def say_text(text, voice_index=0): with open("speech.mpeg", "wb") as f: f.write(response.content) playsound("speech.mpeg") - # Delete audio file os.remove("speech.mpeg") + return True else: print("Request failed with status code:", response.status_code) print("Response content:", response.content) + return False + +def gtts_speech(text): + tts = gtts.gTTS(text) + tts.save("speech.mp3") + playsound("speech.mp3") + os.remove("speech.mp3") + +def say_text(text, voice_index=0): + if not cfg.elevenlabs_api_key: + gtts_speech(text) + else: + success = eleven_labs_speech(text) + if not success: + gtts_speech(text) +