change model

rebuild trigger

gemini flash image edit

upload edited images to imgbb

fix image generation failing

re edit images

handle document images

uptade reply handeling

update debugs

debug

edit image

Increase backoff duration for image fetching and limit returned prompt to first paragraph

Update image generation message to handle singular and plural cases

Improve image generation error handling and update message for image count in webhook

Remove polling functionality and related intent models from the bot

Remove redundant acceptance message in voice reply function and log failure instead

Refactor route_intent function to enhance system prompt clarity and update command descriptions; add music generation functionality

Refactor audio generation prompt handling to remove system prompt from URL encoding

Refactor system prompt formatting and improve command list clarity in configuration

Update system prompt for improved tone and clarity; adjust assistant's personality description

Update LLM model in configuration and streamline model retrieval in code

Enhance audio reply generation with error handling; update assistant's personality in system prompt

Add health check endpoint and improve logging context; update requirements for telegram bot support

Refactor route_intent and whatsapp_webhook functions for improved error handling and logging; streamline JSON parsing and payload validation

Refactor route_intent function to enhance prompt clarity and user interaction; streamline function invocation instructions and JSON response format

Refactor LLM integration and function call handling; streamline message processing and enhance error handling

Refactor LLM integration and function call handling; update default model and streamline message processing

Refactor route_intent function for improved prompt clarity and intent handling; enhance error messaging and logging

Refactor route_intent function for improved clarity and efficiency in prompt handling and intent parsing

Refine sys_prompt in route_intent function for clearer function invocation instructions and improved user interaction

Enhance route_intent function prompt for improved user interaction and clarity on JSON response format

Update default LLM model to Nvidia Llama-3_3-Nemotron-Super-49B-v1

Update route_intent function documentation for clarity on JSON response format

Update user input prompt format in route_intent function for clarity

Fix OpenAI client initialization by setting the api_key to the environment variable

Update LLM model and API configuration; modify example prompt for testing

Update /gen command documentation to correct image size options for clarity.

Remove max_tokens parameter from LLM request in generate_llm function for cleaner API call.

Increase conversation history limit to 20 messages; add summarize, translate, meme, and poll functionalities with corresponding intents and command support in the configuration.

Revert "Refactor intent handling by removing unused actions and improving error handling for LLM requests; update comments for clarity."

Refactor intent handling by removing unused actions and improving error handling for LLM requests; update comments for clarity.

Refactor LLM functions by removing unused summarize, translate, and meme functions; update generate_llm to increase max_tokens limit for improved response handling.

Refactor intent classes by removing unused poll-related intents; update system prompt in config for clarity and conciseness.

Refactor conversation history to limit stored messages to 10; remove unused poll functions and update command guidance in config for clarity and conciseness.

Refactor image and voice reply generation functions for improved readability; enhance logging for better debugging and error handling.

Refactor voice reply prompt to enhance tone; adjust wording for a more friendly and playful delivery.

Refactor error handling in generate_llm to raise LLMBadRequestError on HTTP 400; add clear_history function to reset conversation history on bad requests.

Add user feedback for image generation and enhance command guidance in config

Refactor generate_llm function for improved readability; streamline parameter assignments and enhance error handling for 400 Bad Request responses.

Refactor voice reply generation to include system prompt customization; add helper function to build system prompt and update encoded prompt formatting.