1). Add code for defining temperature, top_p, n, max_tokens in main_config.

2). This will allow users to play around with these values and choose ones that work best for them.

3). This can be defined in main_config file and will not require code modifications.

alwrity AI Tool Development Insights

Overview:

The primary focus during the development of our AI tool, alwrity, was to create a functional and effective AI writer. While the tool is operational, its usability is currently suboptimal.

Development Process:

Considerable time was invested in integrating various libraries and establishing workflows to ensure the tool's functionality. However, this focus on technical aspects led to a neglect of user experience and overall quality.

Challenges Encountered:

There was a notable deficiency in the quality of both prompts and generated content. Prompt quality suffered due to inadequate engineering, and the resulting content failed to meet expectations.

Future Improvements:

Moving forward, our priority is to enhance the usability and effectiveness of alwrity. This includes:

• Prompt Enhancement: Significantly improving the quality and relevance of prompts to elicit better content from the AI.
• Tool Optimization: Making better use of Tavily, Exa, and other existing resources to refine the tool's performance.
• Prompt Engineering: Investing resources in improving prompt engineering processes to drive better content generation.
• Content Quality: Focusing on delivering high-quality, long-form content while maintaining a commitment to avoid generating subpar outputs.

Conclusion:

The current state of alwrity reflects the challenges inherent in the development process. However, with a dedicated effort to address usability issues and refine content generation, we aim to elevate the tool to meet user expectations and deliver exceptional results.

Thanks for the powerful AI-Writer (Alwrity) tool.. we have been using the application over a month now and its free and has no sign up AI writer.
Under 'Include_domains', could you improve the Example and spacing in URL egs.

Adding code to make alwrity multilingual.
The 'blog_language' can be defined in main_config

Experimenting with Gemini multimodal features.
1). Present scope is to include an AI writer capable of inputting local image files and creating either a Blog Post, Alt Text, Marketing Copy Or Food Recipes.

Hi, @AJaySi

Could we add a reference to our Alwrity Essay Writer tool features? It'd be awesome for our users. Thanks!

AI Essay Writer - Alwrity feature

Essay Types and Education Levels

Essay Types

• Argumentative: Forming an opinion via research. Building an evidence-based argument.
• Expository: Knowledge of a topic. Communicating information clearly.
• Narrative: Creative language use. Presenting a compelling narrative.
• Descriptive: Creative language use. Describing sensory details.

Education Levels

• Primary School
• High School
• College
• Graduate School

Inputs and Details

Essay Type

Choose the type of essay you want to write from the options provided.

Education Level

Select your current level of education from the given choices.

Essay Title

Enter the title of your essay. The title should accurately reflect the content and focus of your essay.

Number of Pages

Select the length of your essay by choosing one or more options:

• Short Form (1-2 pages)
• Medium Form (3-5 pages)
• Long Form (6+ pages)

There is a way to generate long articles?
Even if I change the number of tokens, the posts are about 500 words. (GPT or Gemini Pro)

I used on main_config:
model = gpt-3.5-turbo-16k
max_tokens = 8192

Thanks!

I am making changes in the code to pick up values defined in main_config.
Presently, its hard coded and main_config will make it easier to choose/modify behavior of program.
WIP.

Keyword web research is not working. I am getting the below error message.

Tweak the alwrity web researcher to scrape latest news on given keywords.

Google gemini latest feature is system prompts. This should make model adhere more strictly to alwrity's prompts.
Test and check.

https://ai.google.dev/docs/system_instructions/

hello,
# python blogen.py File "blogen.py", line 7 from rich import print ^ SyntaxError: invalid syntax

There is demand for AI copywriting tools.
There are web apps for the same. I will integrate them into alwrity cli.

I tried to generate a blog using audio to blog (Alwrity features). I have seen different outputs (example) in the "wiki". And I am getting a different output. I have attached the output image below. Please help regarding this issue. Thank you.

ERROR|metaphor_basic_neural_web_search.py:213:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\Dell\AI-Blog-Writer\workspace\web_research_reports\kanamo_peak_trek_in_details_2024-03-23_18-41-33'
INFO|combine_research_and_blog.py:75:blog_with_research| Calling OpenAI LLM.
2024-03-23 18:44:14,373-INFO-_client-1013-HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 400 Bad Request"
2024-03-23 18:44:14,388-ERROR-openai_chat_completion-53-OpenAI API Error: Error code: 400 - {'error': {'message': "This model's maximum context length is 16385 tokens. However, your messages resulted in 25664 tokens. Please reduce the length of the messages.", 'type': 'invalid_request_error', 'param': 'messages', 'code': 'context_length_exceeded'}}
PS C:\Users\Dell\AI-Blog-Writer>

1. I am getting the above issue while using open AI API. In Gemini API I do not have the same issue.
2. While using Open AI API, the blog category is not coming in the results, while with Gemini API it's coming.

Thank you for your help regarding this issue.

Our search apis have some options for language. Given the language in main_config, the gpt_online_researcher code should be modified to do search in that language and that locale.

All API calls are not private. Privacy can be supported by running the LLMs locally on your laptop.
This ensures nothing from your laptop goes out of the laptop..

Ollama integration with Alwrity is pending for sometime now. Ollama helps run LLMs locally on your laptop and hence no API calls Or transfer of data to outside world.

Also, this will truly make alwrity Privacy focused and Free. As alwrity will use locally hosted LLMs, it free its users from LLM subscription models and monthly payouts.

https://ollama.com/

1). Write wiki to explain main_config options.
2). Make it simple enough for non-programmers and all users.
3). Expose more options to end-user, remove hard-codings.
4). Explain how-to change prompts.

Drafts:
https://github.com/AJaySi/AI-Blog-Writer/wiki

Hi AJ,
i am unable to save the blog to file. kindly help

Standing on shoulders of Giants.

Assembly transcription, translation with universal1 is at par or better than whisper model. Alwrity will provide audio files, augument them with latest web search and transform into blog content.

Assembly AI provides STT and many more features. They are generous enough with their Free plan.
Alwrity supports openai whisper model and assembly AI will be a great additions for more features.

Hi, Stability AI's latest models need to be supported.

Thank you for your help.

1). Add support for AI story writer or generator.
2). It should produce long form content, more than 5k words atleast.
3). Utilise iterative generation and prompt chaining.
https://github.com/google-gemini/cookbook/blob/main/examples/Story_Writing_with_Prompt_Chaining.ipynb

New Alwrity Feature discussion:

1). Alwrity needs to remember the content and its metadata to not rewrite the same articles again.
Usecase: Each web research can be embedded and stored for alwrity to train itself and find semantically correct answers.
This will help cutoff cost, as we can check local knowledge before calling APIs.

2). This will make alwrity aware of each user preference and also RAG-tune based on their domain.
Making alwrity domain expert for each content type, every user will use it for.
Usecase: If Mr. X uses alwrity to write on marijuana, then alwrity will smoke/store each keywords web research and train itself on it. With each content generation, alwrity will also become domain expert on it.

3). Privacy & Security : Keep your data, Alwrity ain't interested.
Your web searches, content reside at your local disk. The database will also reside on your laptop.

4). It will be interesting to see if alwrity can provide internal linking while content generation.

Exploring : https://github.com/lancedb/lancedb

After cloning the latest code update, I attempted to run it but encountered an issue (mentioned below) where the blog post generation was failing. Could you kindly assist me in troubleshooting this matter? Your help is greatly appreciated. Thank you.

ERROR|google_serp_search.py:309:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Windows\System32\AI-Blog-Writer\AI-Writer\workspace\web_research_reports\google_guideline_in_ai_contents_2024-04-08_18-29-21'
ERROR|google_serp_search.py:309:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Windows\System32\AI-Blog-Writer\AI-Writer\workspace\web_research_reports\google_guideline_in_ai_contents_2024-04-08_18-29-21'
INFO|blog_from_google_serp.py:36:write_blog_google_serp| Generating blog and FAQs from Google web search results.
ERROR|main_text_generation.py:150:read_llm_parameters| An unexpected error occurred: 'charmap' codec can't decode byte 0x9d in position 2087: character maps to
ERROR|main_text_generation.py:55:llm_text_gen| Failed to read LLM parameters: 'charmap' codec can't decode byte 0x9d in position 2087: character maps to
ERROR|blog_from_google_serp.py:41:write_blog_google_serp| Exit: Failed to get response from LLM: 'charmap' codec can't decode byte 0x9d in position 2087: character maps to

Discussed in #52

Exa.ai provides similar search options as detailed here: https://exa.ai/blog/announcing-exa

Presently, these values are hard-coded.
Expose Alwrity search engine parameters in main_config

Discussed in #6

Include streaming output, instead of waiting with no feedbacks.
Include code for all LLMs(presently openai, google)

This is based on a team of agents working together to come up with content calender for given keywords.
One can define the timeframe of the content calender as 2 months etc.

One of the agents will do web research, another one will do google trends analysis for given keywords.
Out of this context, the content calender will created by 2 other agents.

Initial results are encouraging. I am plugging in more tools and provide more concrete web search, keywords etc.

This will help users to brainstorm content ideas. This is basic, but does a good job if you do not know what titles, topics to write.

Disclaimer: To err is Human & AI..

ERROR|google_serp_search.py:312:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: '/home/runner/AI-Blog-Writer/workspace/web_research_reports/Top_5_Scuba_Diving_in_India_2024-03-06_12-09-35'
ERROR|google_serp_search.py:312:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: '/home/runner/AI-Blog-Writer/workspace/web_research_reports/Top_5_Scuba_Diving_in_India_2024-03-06_12-09-35'
INFO|blog_from_google_serp.py:36:write_blog_google_serp| Generating blog and FAQs from web search result.
INFO|blog_from_google_serp.py:46:write_blog_google_serp| Calling OpenAI LLM.
2024-03-06 12:09:47,832-INFO-_client-1013-HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 401 Unauthorized"
2024-03-06 12:09:47,833-ERROR-openai_chat_completion-53-OpenAI API Error: Error code: 401 - {'error': {'message': 'Incorrect API key provided: UhQHYK5O************************************vaM6. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}

I am getting this error. Thank you for your help.

Essay writer tool is not working. The tool page is showing the error given below. Thank you for the help.

ModuleNotFoundError: This app has encountered an error. The original error message is redacted to prevent data leaks. Full error details have been recorded in the logs (if you're on Streamlit Cloud, click on 'Manage app' in the lower right of your app).
Traceback:
File "/home/adminuser/venv/lib/python3.11/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 584, in _run_script
exec(code, module.dict)
File "/mount/src/alwrity-essay/essay_writer.py", line 4, in
import openai

Hi, I am getting the following error while using the Gemini LLM model API.

Thank you for your help.

1. Under Working Agent mention - Agent Content Team identity like researcher, outliner, writer, reviewer

2. If Below error, then Aborted search with message.
   

Alwrity is giving 4 amazing blogs. I am getting errors while saving and writing files. The error messages are below:

ERROR|google_serp_search.py:312:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'
ERROR|google_serp_search.py:312:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'

ERROR|tavily_ai_search.py:155:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'

ERROR|tavily_ai_search.py:155:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'

ERROR|metaphor_basic_neural_web_search.py:213:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'

ERROR|google_trends_researcher.py:490:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'
Silhouette Score: 0.10566724694995207

ERROR|google_trends_researcher.py:490:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'

ERROR|google_trends_researcher.py:490:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'
ERROR|google_trends_researcher.py:529:do_google_trends_analysis| Failed to get results from google trends related topics: The truth value of a DataFrame is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().
ERROR|google_trends_researcher.py:490:save_in_file| Error occurred while writing to the file: [Errno 2] No such file or directory: 'C:\Users\this\AI-Blog-Writer\workspace\web_research_reports\Scuba_Diving_in_Goa_2024_Comprehensive_Guide_2024-03-09_21-46-08'

1). Play around with temperature, top_p, frequency_penalty etc.
2). Do not use most common AI words like:

It’s important to note', 'Delve into', 'Tapestry', 'Bustling', 'In summary', 'In conclusion', 'Remember that', 'Take a dive into', 'Navigating', 'Navigating the landscape', 'Navigating the complexities of', 'Landscape', 'The landscape of', 'Testament', 'a testament to', 'In the world of', 'Realm', 'Embark', 'virtuoso', 'Let's explore', 'symphony', 'Harnessing', 'Revolutionizing', 'Empower', 'game changing', 'ever-changing', 'Embrace', 'Embracing', 'game-changing', 'ever-evolving', etc.

Alwrity Content Calendar for AI Writers

Month 1

Month 2

Note: Content will be optimized to position Alwrity.com as an open-source AI writer, showcasing its unique features and benefits compared to other available tools.

I tried to generate a long-form blog article but encountered an error mentioned below. Thank you for your help.

E:\CHATGPT-AI-Marketing-Tools-CMD\AI-Blog-Writer>py blogen.py
Traceback (most recent call last):
File "E:\CHATGPT-AI-Marketing-Tools-CMD\AI-Blog-Writer\blogen.py", line 6, in
from PyInquirer import prompt
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\PyInquirer_init_.py", line 6, in
from prompt_toolkit.token import Token
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit_init_.py", line 16, in
from .interface import CommandLineInterface
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit\interface.py", line 19, in
from .application import Application, AbortAction
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit\application.py", line 8, in
from .key_binding.bindings.basic import load_basic_bindings
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit\key_binding\bindings\basic.py", line 9, in
from prompt_toolkit.renderer import HeightIsUnknownError
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit\renderer.py", line 11, in
from prompt_toolkit.styles import Style
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit\styles_init_.py", line 8, in
from .from_dict import *
File "D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\site-packages\prompt_toolkit\styles\from_dict.py", line 9, in
from collections import Mapping
ImportError: cannot import name 'Mapping' from 'collections' (D:\Users\vaan1\AppData\Local\Programs\Python\Python312\Lib\collections_init_.py)

E:\CHATGPT-AI-Marketing-Tools-CMD\AI-Blog-Writer>pip blogen.py
ERROR: unknown command "blogen.py"

I have this issue and I don't see the file .env.
What should I do?

Thank you for your help

after i command

!python alwrity.py

this error occurs

NameError: name 'radiolistbox' is not defined

it's a great job, but I can't get it to work. The first versions I was able to create some posts. Although it asks me to create the web_research_reports folder. Sorry for the inconvenience.

🔝 Top: The most popular search topics.
🚀 Rising: Topics experiencing a significant increase in search frequency since the last time period. Topics marked :pile_of_poop:'Breakout' had a tremendous surge, likely because they are new and had few prior searches.
Empty DataFrame
Columns: []
Index: []
INFO|google_trends_researcher.py:488:save_in_file| Search content saved to C:\Flota\AI-Blog-Writer\workspace\web_research_reports\los_mejores_quesos_del_mundo_2024-04-10_16-31-23
ERROR|google_trends_researcher.py:529:do_google_trends_analysis| Failed to get results from google trends related topics: The truth value of a DataFrame is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().
ERROR|google_trends_researcher.py:490:save_in_file| Error occurred while writing to the file: 'charmap' codec can't encode characters in position 0-207: character maps to
╒════╤═══════════════════════════════════════════════════╤═════════════════════════════════════════════╤══════════════════════════════════════════════════════╤════════════════════════════════════════════════╕
│ │ K📢eyword Col1 │ K📢eyword Col2 │ K📢eyword Col3 │ K📢eyword Col4 │
╞════╪═══════════════════════════════════════════════════╪═════════════════════════════════════════════╪══════════════════════════════════════════════════════╪════════════════════════════════════════════════╡
│ 0 │ ranking de los mejores quesos del mundo │ los mejores quesos de todo el mundo │ que or de │ los mejores quesos del mundo 2024 │
├────┼───────────────────────────────────────────────────┼─────────────────────────────────────────────┼──────────────────────────────────────────────────────┼────────────────────────────────────────────────┤
│ 1 │ cuales son los mejores quesos del mundo │ lista de los mejores quesos del mundo │ el mejor queso del mundo │ mejores quesos del mundo lista │
├────┼───────────────────────────────────────────────────┼─────────────────────────────────────────────┼──────────────────────────────────────────────────────┼────────────────────────────────────────────────┤
│ 2 │ donde estan los mejores quesos del mundo │ donde hacen los mejores quesos del mundo │ donde se hacen los mejores quesos del mundo │ de donde son los mejores quesos del mundo │
├────┼───────────────────────────────────────────────────┼─────────────────────────────────────────────┼──────────────────────────────────────────────────────┼────────────────────────────────────────────────┤
│ 3 │ como vs que comparison │ donde producen los mejores quesos del mundo │ en que pais se producen los mejores quesos del mundo │ donde se producen los mejores quesos del mundo │
├────┼───────────────────────────────────────────────────┼─────────────────────────────────────────────┼──────────────────────────────────────────────────────┼────────────────────────────────────────────────┤
│ 4 │ en donde se producen los mejores quesos del mundo │ what is del mundo │ what does del mundo mean │
│
╘════╧═══════════════════════════════════════════════════╧═════════════════════════════════════════════╧══════════════════════════════════════════════════════╧════════════════════════════════════════════════╛
INFO|keywords_to_blog.py:73:write_blog_from_keywords| Google Trends keywords to use in the blog: ['ranking de los mejores quesos del mundo', ' los mejores quesos de todo el mundo', ' que or de', ' los mejores quesos del mundo 2024', ' cuales son los mejores quesos del mundo', ' lista de los mejores quesos del mundo', ' el mejor queso del mundo', ' mejores quesos del mundo lista', ' donde estan los mejores quesos del mundo', ' donde hacen los mejores quesos del mundo', ' donde se hacen los mejores quesos del mundo', ' de donde son los mejores quesos del mundo', ' como vs que comparison', ' donde producen los mejores quesos del mundo', ' en que pais se producen los mejores quesos del mundo', ' donde se producen los mejores quesos del mundo', ' en donde se producen los mejores quesos del mundo', ' what is del mundo',ERROR|main_text_generation.py:150:read_llm_parameters| An unexpected error occurred: 'charmap' codec can't decode byte 0x9d in position 2106: character maps to
ERROR|main_text_generation.py:55:llm_text_gen| Failed to read LLM parameters: 'charmap' codec can't decode byte 0x9d in position 2106: character maps to
ERROR|combine_blog_and_keywords.py:30:blog_with_keywords| blog_with_keywords: Failed to get response from LLM: 'charmap' codec can't decode byte 0x9d in position 2106: character maps to
ERROR|keywords_to_blog.py:76:write_blog_from_keywords| Failed to do Google Trends Analysis:'charmap' codec can't decode byte 0x9d in position 2106: character maps to
INFO|keywords_to_blog.py:77:write_blog_from_keywords| ########### Blog Content After Google Trends Analysis:#########

INFO|keywords_to_blog.py:85:write_blog_from_keywords| Pass Final blog for blog-proof reading and *improvements.
ERROR|main_text_generation.py:150:read_llm_parameters| An unexpected error occurred: 'charmap' codec can't decode byte 0x9d in position 2106: character maps to
ERROR|main_text_generation.py:55:llm_text_gen| Failed to read LLM parameters: 'charmap' codec can't decode byte 0x9d in position 2106: character maps to
Failed to write blog on los mejores quesos del mundo, Error: name 'logger' is not defined

Integrate/plug-in code for AI FAQ generator.

LLMs restrict output tokens/length of text generation. The free open do so even more, hindering generation of long for articles.
Prompt chaining and iterative generation will help us overcome this limitation. The input context window is large, thus we can input a lot and get short outputs(llm context window).

Usecases of Writing long-form content:

Break down the writing process into outlined sections or chapters. Provide prompts for each section, allowing the AI to expand upon them sequentially, ensuring a coherent and structured narrative.

Research projects:

Guide the AI through the research process by prompting it to:

1. Find relevant source documents or materials.
2. Extract key facts and data from the sources.
3. Synthesize conclusions or insights based on the gathered information, ensuring accuracy and relevance.

Data analysis:

Facilitate the data analysis process by instructing the AI to:

1. Import datasets from various sources.
2. Clean and preprocess the data to ensure its quality and consistency.
3. Run appropriate analyses or algorithms to derive meaningful insights.
4. Generate visualizations such as charts or graphs to illustrate the findings effectively.
5. Interpret and summarize the results in a clear and concise manner, providing actionable conclusions.

Computer programming:

Break down programming tasks into manageable steps, guiding the AI to:

1. Outline the logic and functionality of the program, specifying inputs, processes, and outputs.
2. Write pseudocode to represent the algorithmic steps in a human-readable format.
3. Translate the pseudocode into actual code using the chosen programming language or framework.
4. Debug and troubleshoot any errors or issues encountered during the coding process, ensuring the program's correctness and efficiency.

Travel planning:

Assist in planning trips by prompting the AI to:

1. Suggest destinations based on user-defined criteria such as budget, interests, or preferences.
2. Find suitable flights, accommodations, and other travel-related information, considering factors like availability and pricing.
3. Create a detailed daily itinerary outlining activities, attractions, and dining options for each day of the trip.
4. Generate a comprehensive packing list based on the planned activities and duration of the journey, ensuring travelers are well-prepared for their adventure.

Hello,

I'm currently using OpenAI ChatGPT 3.5 Turbo in Alwrity. I'd like to switch to the Gemini LLM model. Could you please guide me through the steps involved in making this change?

Thank you for your assistance.

Continual in-context generation should be the future as LLM context windows will tend to become infinite.

Crewai content team is present in alwrity to build upon. There dire need to simplify it for people not used to *.py files.
Crewai content planning & calender is giving satisfactory results.

Alwrity uses Tavily(gpt-researcher) for providing context for content generation.

This suggestions seem worth implementing with Alwrity, and provide it as an option for people to choose from.
https://github.com/assafelovic/gpt-researcher/tree/master/multi_agents

Alwrity will encapsulate all the complexities, for non tech content creators.