Technical documentation - Yacana 0.2.0

OllamaAgent

▶️ Member variables

Name	Type	Description
`name`	str	Name of the agent. Can be used during conversations. Use something short and meaningful that doesn't contradict the system prompt.
`model_name`	str	Name of the LLM model that will be sent to the inference server. For instance 'llama:3.1' or 'mistral:latest' etc.
`system_prompt`	str \| None	Defines the way the LLM will behave. For instance set the SP to "You are a pirate" to have it talk like a pirate.
`model_settings`	ModelSettings	All settings that Ollama currently supports as model configuration. This needs to be tested with other inference servers. This allows modifying deep behavioral patterns of the LLM.
`headers`	dict	Custom headers for requests.
`endpoint`	str \| None	Endpoint URL for the inference server.
`runtime_config`	Dict	Runtime configuration for the agent.
`task_runtime_config`	Dict	Runtime configuration for tasks.
`history`	History	The conversation history.

▶️ Methods

➡️ init(...)

Initialize a new OllamaAgent instance.

Parameter name	Parameter type	Parameter description
`name`	str	Name of the agent. Use something short and meaningful that doesn't contradict the system prompt.
`model_name`	str	Name of the LLM model that will be sent to the inference server (e.g., 'llama:3.1' or 'mistral:latest').
`system_prompt`	str \| None	Defines the way the LLM will behave (e.g., "You are a pirate" to have it talk like a pirate). Defaults to None.
`endpoint`	str	The Ollama endpoint URL. Defaults to "http://127.0.0.1:11434".
`headers`	dict	Custom headers to be sent with the inference request. Defaults to None.
`model_settings`	OllamaModelSettings	All settings that Ollama currently supports as model configuration. Defaults to None.
`runtime_config`	Dict \| None	Runtime configuration for the agent. Defaults to None.

Return type: OllamaAgent

➡️ export_to_file(...)

Exports the current agent configuration to a file. This contains all the agents data and history. This means that you can use the import_from_file method to load this agent back again and continue where you left off.

Parameter name	Parameter type	Parameter description
`file_path`	str	Path of the file in which to save the data. Specify the path + filename. Be wary when using relative paths.
`strip_api_token`	bool	If True, removes the API token from the exported data. Defaults to False.
`strip_headers`	bool	If True, removes headers from the exported data. Defaults to False.

Return type: None

➡️ import_from_file(...)

Loads the state previously exported from the export_to_file method. This will return an Agent in the same state as it was before it was saved, allowing you to resume the agent conversation even after the program has exited.

Parameter name	Parameter type	Parameter description
`file_path`	str	The path to the file from which to load the Agent.

Return type: OllamaAgent

OpenAiAgent

▶️ Member variables

Name	Type	Description
`name`	str	Name of the agent. Use something short and meaningful that doesn't contradict the system prompt.
`model_name`	str	Name of the LLM model that will be sent to the inference server (e.g., 'gpt-4' or 'gpt-3.5-turbo').
`system_prompt`	str \| None	Defines the way the LLM will behave (e.g., "You are a pirate" to have it talk like a pirate).
`model_settings`	OpenAiModelSettings	All settings that OpenAI currently supports as model configuration.
`api_token`	str	The API token for authentication.
`headers`	dict	Custom headers to be sent with the inference request.
`endpoint`	str \| None	The OpenAI endpoint URL.
`runtime_config`	Dict	Runtime configuration for the agent.
`task_runtime_config`	Dict	Runtime configuration for tasks.
`history`	History	The conversation history.

▶️ Methods

➡️ init(...)

Initialize a new OpenAiAgent instance.

Parameter name	Parameter type	Parameter description
`name`	str	Name of the agent. Use something short and meaningful that doesn't contradict the system prompt.
`model_name`	str	Name of the LLM model that will be sent to the inference server (e.g., 'gpt-4' or 'gpt-3.5-turbo').
`system_prompt`	str \| None	Defines the way the LLM will behave (e.g., "You are a pirate" to have it talk like a pirate). Defaults to None.
`endpoint`	str \| None	The OpenAI endpoint URL. Defaults to None (uses OpenAI's default endpoint).
`api_token`	str	The API token for authentication. Defaults to an empty string.
`headers`	dict	Custom headers to be sent with the inference request. Defaults to None.
`model_settings`	OpenAiModelSettings	All settings that OpenAI currently supports as model configuration. Defaults to None.
`runtime_config`	Dict \| None	Runtime configuration for the agent. Defaults to None.

Return type: OpenAiAgent

➡️ export_to_file(...)

Parameter name	Parameter type	Parameter description
`file_path`	str	Path of the file in which to save the data. Specify the path + filename. Be wary when using relative paths.
`strip_api_token`	bool	If True, removes the API token from the exported data. Defaults to False.
`strip_headers`	bool	If True, removes headers from the exported data. Defaults to False.

Return type: None

➡️ import_from_file(...)

Parameter name	Parameter type	Parameter description
`file_path`	str	The path to the file from which to load the Agent.

Return type: OpenAiAgent

Task

▶️ Member variables

Name	Type	Description
`prompt`	str	The task to solve. It is the prompt given to the assigned LLM.
`agent`	GenericAgent	The agent assigned to this task.
`json_output`	bool	If True, will force the LLM to answer as JSON.
`structured_output`	Type[BaseModel] \| None	The expected structured output type for the task. If provided, the LLM's response will be validated against this type.
`tools`	List[Tool]	A list of tools that the LLM will get access to when trying to solve this task.
`medias`	List[str] \| None	An optional list of paths pointing to images on the filesystem.
`llm_stops_by_itself`	bool	Only useful when the task is part of a GroupSolve(). Signals the assigned LLM that it will have to stop talking by its own means.
`use_self_reflection`	bool	Only useful when the task is part of a GroupSolve(). Allows keeping the self reflection process done by the LLM in the next GS iteration.
`forget`	bool	When True, the Agent won't remember this task after completion. Useful for routing purposes.
`streaming_callback`	Callable \| None	Optional callback for streaming responses.
`runtime_config`	Dict \| None	Optional runtime configuration for the task.
`tags`	List[str] \| None	Optional list of tags that will be added to all message(s) corresponding to this task.

▶️ Methods

➡️ init(...)

Initialize a new Task instance.

Parameter name	Parameter type	Parameter description
`prompt`	str	The task to solve. It is the prompt given to the assigned LLM.
`agent`	GenericAgent	The agent assigned to this task.
`json_output`	bool	If True, will force the LLM to answer as JSON. Defaults to False.
`structured_output`	Type[BaseModel] \| None	The expected structured output type for the task. If provided, the LLM's response will be validated against this type. Defaults to None.
`tools`	List[Tool]	A list of tools that the LLM will get access to when trying to solve this task. Defaults to an empty list.
`medias`	List[str] \| None	An optional list of paths pointing to images on the filesystem. Defaults to None.
`llm_stops_by_itself`	bool	Only useful when the task is part of a GroupSolve(). Signals the assigned LLM that it will have to stop talking by its own means. Defaults to False.
`use_self_reflection`	bool	Only useful when the task is part of a GroupSolve(). Allows keeping the self reflection process done by the LLM in the next GS iteration. Defaults to False.
`forget`	bool	When True, the Agent won't remember this task after completion. Useful for routing purposes. Defaults to False.
`streaming_callback`	Callable \| None	Optional callback for streaming responses. Defaults to None.
`runtime_config`	Dict \| None	Optional runtime configuration for the task. Defaults to None.
`tags`	List[str] \| None	Optional list of tags that will be added to the message(s) corresponding to the user's prompt. Defaults to None.

Return type: Task

➡️ add_tool(...)

Add a tool to the list of tools available for this task.

Parameter name	Parameter type	Parameter description
`tool`	Tool	The tool to add to the task's tool list.

Return type: None

➡️ solve()

Execute the task using the assigned LLM agent. This method will call the assigned LLM to perform inference on the task's prompt. If tools are available, the LLM may use them, potentially making multiple calls to the inference server.
Return type: Message

▶️ Properties

➡️ uuid

Get the unique identifier for this task.
Return type: Message

History

▶️ Member variables

Name	Type	Description
`slots`	List[HistorySlot]	List of history slots.

▶️ Methods

➡️ add_slot(...)

Adds a new slot to the history at the specified position.

Parameter name	Parameter type	Parameter description
`history_slot`	HistorySlot	The slot to add to the history.
`position`	int \| SlotPosition	The position where to add the slot. Can be an integer or a SlotPosition enum value. Defaults to SlotPosition.BOTTOM.

Return type: None

➡️ delete_slot(...)

Deletes a slot from the history.

Parameter name	Parameter type	Parameter description
`index`	int	The index of the slot to delete.

Return type: None

➡️ get_last_slot()

Returns the last slot of the history. A good syntactic sugar to get the last item from the conversation.

Return type: HistorySlot

➡️ get_slot_by_index(...)

Returns the slot at the given index.

Parameter name	Parameter type	Parameter description
`index`	int	The index of the slot to return.

Return type: HistorySlot

➡️ get_slot_by_id(...)

Returns the slot with the given ID.

Parameter name	Parameter type	Parameter description
`id`	str	The ID of the slot to return.

Return type: HistorySlot

➡️ get_slot_by_message(...)

Returns the slot containing the given message.

Parameter name	Parameter type	Parameter description
`message`	GenericMessage	The message to search for.

Return type: HistorySlot

➡️ add_message(...)

Adds a new message to the history by creating a new slot.

Parameter name	Parameter type	Parameter description
`message`	GenericMessage	The message to add to the history.

Return type: None

➡️ get_messages_as_dict()

Returns all messages in the history as a list of dictionaries.

Return type: List[Dict]

➡️ pretty_print()

Prints the history to stdout with colored output.

Return type: None

➡️ create_check_point()

Creates a checkpoint of the current history state.

Return type: str

➡️ load_check_point(...)

Loads a checkpoint of the history. Perfect for a timey wimey rollback in time.

Parameter name	Parameter type	Parameter description
`uid`	str	The unique identifier of the checkpoint to load.

Return type: None

➡️ get_message(...)

Returns the message at the given index.

Parameter name	Parameter type	Parameter description
`index`	int	The index of the message to return.

Return type: GenericMessage

➡️ get_messages_by_tags(...)

Returns messages that match the given tags based on the matching mode.

Parameter name	Parameter type	Parameter description
`tags`	List[str]	The tags to filter messages by.
`strict`	bool	Controls the matching mode. If False (default), returns messages that have ANY of the specified tags. If True, returns messages that have EXACTLY the specified tags (and possibly more). Defaults to False.

Return type: Sequence[GenericMessage]

➡️ get_last_message()

Returns the last message in the history.

Return type: GenericMessage

➡️ get_all_messages()

Returns all messages in the history.

Return type: List[Message]

➡️ clean()

Resets the history, preserving only the initial system prompt if present.

Return type: None

HistorySlot

▶️ Member variables

Name	Type	Description
`id`	str	The unique identifier for the slot.
`creation_time`	int	The timestamp when the slot was created.
`messages`	List[GenericMessage]	List of messages in the slot.
`raw_llm_json`	str \| None	The raw LLM JSON response for the slot.
`main_message_index`	int	The index of the main message in the slot.

▶️ Methods

➡️ init(...)

Initialize a new HistorySlot instance.

Parameter name	Parameter type	Parameter description
`messages`	List[GenericMessage] \| None	A list of messages. Each message is a variation of the main message defined by the @main_message_index parameter. Defaults to None.
`raw_llm_json`	str \| None	The raw LLM JSON response for the slot. This is the raw JSON from the inference server. When using OpenAI this may contain more than one message hence the slot system acts as a container for the messages. Defaults to None.

Return type: HistorySlot

➡️ set_main_message_index(...)

A slot can contain any number of concurrent messages. But only one can be the main slot message and actually be part of the History. This method sets the index of the main message within the list of available messages in the slot.

Parameter name	Parameter type	Parameter description
`message_index`	int	The index of the message to select as the main message.

Return type: None

➡️ get_main_message_index()

Returns the index of the main message in the slot.

Return type: int

➡️ add_message(...)

Adds a new message to the slot.

Parameter name	Parameter type	Parameter description
`message`	GenericMessage	The message to add to the slot.

Return type: None

➡️ get_message(...)

Returns the main message of the slot or the one at the given index if index is provided.

Parameter name	Parameter type	Parameter description
`message_index`	int \| None	The index of the message to return. If None, returns the currently selected message.

Return type: GenericMessage

➡️ get_all_messages()

Returns all the messages in the slot.

Return type: List[GenericMessage]

➡️ set_raw_llm_json(...)

Sets the raw LLM JSON response for the slot.

Parameter name	Parameter type	Parameter description
`raw_llm_json`	str	The raw JSON response from the LLM.

Return type: None

➡️ delete_message_by_index(...)

Deletes a message from the slot by index.

Parameter name	Parameter type	Parameter description
`message_index`	int	The index of the message to delete.

Return type: None

➡️ delete_message_by_id(...)

Deletes a message from the slot by id.

Parameter name	Parameter type	Parameter description
`message_id`	str	The ID of the message to delete.

Return type: None

➡️ keep_only_selected_message()

Keeps only the currently selected message in the slot and deletes all the others. If there's only one message, this method does nothing.

Return type: None

Message

▶️ Member variables

Name	Type	Description
`id`	str	The unique identifier of the message.
`role`	MessageRole	From whom is the message from.
`content`	str \| None	The actual message content. Can be None if tool_calls is provided.
`tool_calls`	List[ToolCallFromLLM] \| None	List of tool calls associated with the message.
`medias`	List[str]	List of media file paths.
`structured_output`	Type[T] \| None	Pydantic model for structured output.
`tool_call_id`	str \| None	ID of the associated tool call.
`tags`	List[str]	List of tags associated with the message.

▶️ Methods

➡️ init(...)

Initialize a new Message instance.

Parameter name	Parameter type	Parameter description
`role`	MessageRole	The role of the message sender.
`content`	str	The content of the message.
`tags`	List[str]	Optional list of tags associated with the message. Defaults to None.

Return type: Message

➡️ get_message_as_dict()

Convert the message to a dictionary format.

Return type: dict

➡️ get_as_pretty()

Get a pretty-printed string representation of the message.

Return type: str

➡️ add_tags(...)

Add a tag to the message.

Parameter name	Parameter type	Parameter description
`tags`	List[str]	The tag to add to the message.

Return type: None

➡️ remove_tag(...)

Remove a tag from the message. Tags are used to filter messages in the history.

Parameter name	Parameter type	Parameter description
`tag`	str	The tag to remove from the message.

Return type: None

Tool

▶️ Member variables

Name	Type	Description
`tool_name`	str	A name for the tool. Should be concise and related to what the tool does.
`function_description`	str	A description for the tool. Should be concise and related to what the tool does. May contain an example of how to use. Refer to the documentation.
`function_ref`	Callable	The reference to a python function that will be called with parameters provided by the LLM.
`optional`	bool	Allows to a certain extent the LLM to choose to use the given tool or not depending on the task to solve.
`usage_examples`	List[dict]	A list of python dictionary examples of how the tool should be called. The examples will be given to the LLM to help it call the tool correctly.
`max_custom_error`	int	The max errors a tool can raise. A tool should raise a ToolError(...) exception with a detailed explanation of why it failed.
`max_call_error`	int	The max number of times Yacana can fail to call a tool correctly.

▶️ Methods

➡️ init(...)

Initialize a new Tool instance.

Parameter name	Parameter type	Parameter description
`tool_name`	str	A name for the tool. Should be concise and related to what the tool does.
`function_description`	str	A description for the tool. Should be concise and related to what the tool does. May contain an example of how to use.
`function_ref`	Callable	The reference to a python function that will be called with parameters provided by the LLM.
`optional`	bool	Whether the tool is optional. Defaults to False.
`usage_examples`	List[dict] \| None	Examples of how to use the tool. Defaults to None.
`max_custom_error`	int	Maximum number of custom errors allowed. Defaults to 5.
`max_call_error`	int	Maximum number of call errors allowed. Defaults to 5.

Return type: Tool

OllamaModelSettings

▶️ Member variables

Name	Type	Description
`mirostat`	int \| None	Controls the model's creativity level (0: off, 1: on, 2: extra on).
`mirostat_eta`	float \| None	Adjusts how quickly the model learns from context (e.g., 0.1).
`mirostat_tau`	float \| None	Controls topic adherence (e.g., 5.0).
`num_ctx`	int \| None	Determines context window size (e.g., 4096).
`num_gqa`	int \| None	Controls parallel task handling (e.g., 8).
`num_gpu`	int \| None	Sets GPU utilization (e.g., 50).
`num_thread`	int \| None	Controls parallel processing (e.g., 8).
`repeat_last_n`	int \| None	Controls repetition prevention window (e.g., 64).
`repeat_penalty`	float \| None	Penalty for repeated content (e.g., 1.1).
`temperature`	float \| None	Controls response randomness (e.g., 0.7).
`seed`	int \| None	Random seed for reproducibility (e.g., 42).
`stop`	List[str] \| None	Stop sequences for generation.
`tfs_z`	float \| None	Controls response randomness reduction (e.g., 2.0).
`num_predict`	int \| None	Maximum tokens to generate (e.g., 128).
`top_k`	int \| None	Limits token selection (e.g., 40).
`top_p`	float \| None	Controls token selection probability (e.g., 0.9).

▶️ Methods

➡️ init(...)

Initialize a new OllamaModelSettings instance.

Parameter name	Parameter type	Parameter description
`mirostat`	int \| None	Controls the model's creativity level (0: off, 1: on, 2: extra on). Defaults to None.
`mirostat_eta`	float \| None	Adjusts how quickly the model learns from context (e.g., 0.1). Defaults to None.
`mirostat_tau`	float \| None	Controls topic adherence (e.g., 5.0). Defaults to None.
`num_ctx`	int \| None	Determines context window size (e.g., 4096). Defaults to None.
`num_gqa`	int \| None	Controls parallel task handling (e.g., 8). Defaults to None.
`num_gpu`	int \| None	Sets GPU utilization (e.g., 50). Defaults to None.
`num_thread`	int \| None	Controls parallel processing (e.g., 8). Defaults to None.
`repeat_last_n`	int \| None	Controls repetition prevention window (e.g., 64). Defaults to None.
`repeat_penalty`	float \| None	Penalty for repeated content (e.g., 1.1). Defaults to None.
`temperature`	float \| None	Controls response randomness (e.g., 0.7). Defaults to None.
`seed`	int \| None	Random seed for reproducibility (e.g., 42). Defaults to None.
`stop`	List[str] \| None	Stop sequences for generation. Defaults to None.
`tfs_z`	float \| None	Controls response randomness reduction (e.g., 2.0). Defaults to None.
`num_predict`	int \| None	Maximum tokens to generate (e.g., 128). Defaults to None.
`top_k`	int \| None	Limits token selection (e.g., 40). Defaults to None.
`top_p`	float \| None	Controls token selection probability (e.g., 0.9). Defaults to None.

Return type: OllamaModelSettings

➡️ get_settings()

Get all current settings as a dictionary.

Return type: dict

➡️ reset()

Reset all settings to their initial values.

Return type: None

OpenAiModelSettings

▶️ Member variables

Name	Type	Description
`audio`	Any \| None	Parameters for audio output when using audio modality.
`frequency_penalty`	float \| None	Penalty for token frequency (-2.0 to 2.0).
`logit_bias`	Dict \| None	Token bias adjustments (-100 to 100).
`logprobs`	bool \| None	Whether to return token log probabilities.
`max_completion_tokens`	int \| None	Maximum tokens to generate.
`metadata`	Dict \| None	Additional metadata (max 16 key-value pairs).
`modalities`	List[str] \| None	Output types to generate (e.g., ["text", "audio"]).
`n`	int \| None	Number of completion choices to generate.
`prediction`	Any \| None	Configuration for predicted output.
`presence_penalty`	float \| None	Penalty for token presence (-2.0 to 2.0).
`reasoning_effort`	str \| None	Reasoning effort level ("low", "medium", "high").
`seed`	int \| None	Random seed for reproducibility.
`service_tier`	str \| None	Latency tier for processing ("auto" or "default").
`stop`	str \| List \| None	Stop sequences for generation.
`store`	bool \| None	Whether to store completion output.
`stream_options`	Any \| None	Options for streaming response.
`temperature`	float \| None	Sampling temperature (0 to 2).
`top_logprobs`	int \| None	Number of top tokens to return (0 to 20).
`top_p`	float \| None	Nucleus sampling parameter.
`user`	str \| None	End-user identifier.
`web_search_options`	Any \| None	Web search configuration.

▶️ Methods

➡️ init(...)

Initialize a new OpenAiModelSettings instance.

Parameter name	Parameter type	Parameter description
`audio`	Any \| None	Parameters for audio output when using audio modality. Defaults to None.
`frequency_penalty`	float \| None	Penalty for token frequency (-2.0 to 2.0). Defaults to None.
`logit_bias`	Dict \| None	Token bias adjustments (-100 to 100). Defaults to None.
`logprobs`	bool \| None	Whether to return token log probabilities. Defaults to None.
`max_completion_tokens`	int \| None	Maximum tokens to generate. Defaults to None.
`metadata`	Dict \| None	Additional metadata (max 16 key-value pairs). Defaults to None.
`modalities`	List[str] \| None	Output types to generate (e.g., ["text", "audio"]). Defaults to None.
`n`	int \| None	Number of completion choices to generate. Defaults to None.
`prediction`	Any \| None	Configuration for predicted output. Defaults to None.
`presence_penalty`	float \| None	Penalty for token presence (-2.0 to 2.0). Defaults to None.
`reasoning_effort`	str \| None	Reasoning effort level ("low", "medium", "high"). Defaults to None.
`seed`	int \| None	Random seed for reproducibility.
`service_tier`	str \| None	Latency tier for processing ("auto" or "default").
`stop`	str \| List \| None	Stop sequences for generation.
`store`	bool \| None	Whether to store completion output.
`stream_options`	Any \| None	Options for streaming response.
`temperature`	float \| None	Sampling temperature (0 to 2).
`top_logprobs`	int \| None	Number of top tokens to return (0 to 20).
`top_p`	float \| None	Nucleus sampling parameter.
`user`	str \| None	End-user identifier.
`web_search_options`	Any \| None	Web search configuration.

Return type: OpenAiModelSettings

➡️ get_settings()

Get all current settings as a dictionary.

Return type: dict

➡️ reset()

Reset all settings to their initial values.

Return type: None

EndChatMode

Name	Value	Description
`ALL_TASK_MUST_COMPLETE`	"ALL_TASK_MUST_COMPLETE"	Chat will continue going until all LLMs with @llm_stops_by_itself=True says they are finished. Set precise completion goals in the task prompt if you want this to actually work.
`ONE_LAST_CHAT_AFTER_FIRST_COMPLETION`	"ONE_LAST_CHAT_AFTER_FIRST_COMPLETION"	One agent will have the opportunity to respond after the completion of one agent allowing it to answer one last time.
`ONE_LAST_GROUP_CHAT_AFTER_FIRST_COMPLETION`	"ONE_LAST_GROUP_CHAT_AFTER_FIRST_COMPLETION"	All agents will have one last table turn to speak before exiting the chat after the first completion arrives.
`END_CHAT_AFTER_FIRST_COMPLETION`	"END_CHAT_AFTER_FIRST_COMPLETION"	Immediately stops group chat after an agent has reached completion.
`MAX_ITERATIONS_ONLY`	"MAX_ITERATIONS_ONLY"	Agents won't be asked if they have fulfilled their objectives but instead will loop until achieving max iteration. Max iteration can be set in the EndChat() class.

All types of group chat completion. The difficulty in making agents talk to each other is not to have them talk but to have them stop talking. Note that only tasks that have the @llm_stops_by_itself=True are actually impacted by the mode set here. Use in conjunction with EndChat().

EndChat

▶️ Member variables

Name	Type	Description
`mode`	EndChatMode	The modality to end a chat with multiple agents.
`max_iterations`	int	The max number of iterations in a conversation. An iteration is complete when we get back to the first speaker.

▶️ Methods

➡️ init(...)

Defines the modality of how and when LLMs stop chatting.

Parameter name	Parameter type	Parameter description
`mode`	EndChatMode	The modality to end a chat with multiple agents.
`max_iterations`	int	The max number of iterations in a conversation. An iteration is complete when we get back to the first speaker. Defaults to 5.

Return type: EndChat

GroupSolve

▶️ Member variables

Name	Type	Description
`tasks`	List[Task]	All tasks that must be solved during group chat.
`mode`	EndChatMode	The modality to end a chat with multiple agents.
`reconcile_first_message`	bool	Should the first message from both LLMs be available to one another. Only useful in dual chat.
`max_iter`	int	The max number of iterations in a conversation. An iteration is complete when we get back to the first speaker.
`shift_owner`	Task	The Task to which the shift message should be assigned to. In the end it's rather the corresponding Agent than the Task that is involved here.
`shift_content`	str \| None	A custom message instead of using the opposite agent response as shift message content.

▶️ Methods

➡️ init(...)

Initialize a new GroupSolve instance.

Parameter name	Parameter type	Parameter description
`tasks`	List[Task]	All tasks that must be solved during group chat.
`end_chat`	EndChat	Defines the modality of how and when LLMs stop chatting.
`reconcile_first_message`	bool	Should the first message from both LLMs be available to one another. Only useful in dual chat. Defaults to False.
`shift_message_owner`	Task	The Task to which the shift message should be assigned to. In the end it's rather the corresponding Agent than the Task that is involved here. Defaults to None.
`shift_message_content`	str \| None	A custom message instead of using the opposite agent response as shift message content. Defaults to None.

Return type: GroupSolve

➡️ solve()

Starts the group chat and allows all LLMs to solve their assigned tasks. Note that 'dual chat' and '3 and more' chat have a different way of starting. Refer to the official documentation for more details.

Return type: None

LoggerManager

▶️ Member variables

Name	Type	Description
`LOG_LEVEL`	str	Default log level for the application.
`AVAILABLE_LOG_LEVELS`	list[str]	List of valid log levels that can be set.

▶️ Static Methods

➡️ set_log_level(...)

Set the logging level for the application. This method allows changing the logging level at runtime. All log levels have specific colors, and for INFO level, the color depends on whether it's a user prompt or an AI answer. Setting log_level to None will disable logging.

Parameter name	Parameter type	Parameter description
`log_level`	str \| None	The desired log level. Must be one of: "DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL", or None. If None, logging will be disabled.

Return type: None

➡️ set_library_log_level(...)

Set the logging level for a specific Python library. This is useful to control logging from external libraries independently from the main application logging level.

Parameter name	Parameter type	Parameter description
`library_name`	str	The name of the target library.
`log_level`	str	The desired log level. Must be one of: "DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL", or None.

Return type: None

Exceptions

▶️ ToolError

Exception raised when a tool encounters incorrect parameters. This exception is raised by the user from inside a tool when the tool determines that the given parameters are incorrect. The error message should be didactic to help the LLM fix its mistakes, otherwise it may loop until reaching MaxToolErrorIter.

Parameter name	Parameter type	Parameter description
`message`	str	A descriptive message that will be given to the LLM to help fix the issue. Example: 'Parameter xxxx expected type int but got type string'.

▶️ MaxToolErrorIter

Exception raised when maximum error iterations are reached. This exception is raised when the maximum number of allowed errors has been reached. This includes both errors raised from the tool itself and errors when Yacana fails to call the tool correctly.

Parameter name	Parameter type	Parameter description
`message`	str	Information about which specific iteration counter reached its maximum.

▶️ ReachedTaskCompletion

Exception raised when a task has been successfully completed. This exception is used to signal that the current task has reached its completion state. It is typically used to break out of processing loops or to indicate successful task termination.

▶️ IllogicalConfiguration

Exception raised when the framework is used in an incoherent way. This exception indicates that the framework has been configured or used in a way that doesn't make logical sense. The stacktrace message should provide details about the specific configuration issue.

Parameter name	Parameter type	Parameter description
`message`	str	A description of the illogical configuration or usage.

▶️ TaskCompletionRefusal

Exception raised when the model refuses to complete a task. This exception is raised when the model explicitly refuses to complete the requested task, typically due to ethical, safety, or capability reasons.

Parameter name	Parameter type	Parameter description
`message`	str	The reason for the task completion refusal.

▶️ UnknownResponseFromLLM

Exception raised when the model returns an unknown response.

Parameter name	Parameter type	Parameter description
`message`	str	The unknown response that was received from the LLM.

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ export_to_file(...)

➡️ import_from_file(...)

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ export_to_file(...)

➡️ import_from_file(...)

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ add_tool(...)

➡️ solve()

▶️ Properties

➡️ uuid

▶️ Member variables

▶️ Methods

➡️ add_slot(...)

➡️ delete_slot(...)

➡️ get_last_slot()

➡️ get_slot_by_index(...)

➡️ get_slot_by_id(...)

➡️ get_slot_by_message(...)

➡️ add_message(...)

➡️ get_messages_as_dict()

➡️ pretty_print()

➡️ create_check_point()

➡️ load_check_point(...)

➡️ get_message(...)

➡️ get_messages_by_tags(...)

➡️ get_last_message()

➡️ get_all_messages()

➡️ clean()

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ set_main_message_index(...)

➡️ get_main_message_index()

➡️ add_message(...)

➡️ get_message(...)

➡️ get_all_messages()

➡️ set_raw_llm_json(...)

➡️ delete_message_by_index(...)

➡️ delete_message_by_id(...)

➡️ keep_only_selected_message()

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ get_message_as_dict()

➡️ get_as_pretty()

➡️ add_tags(...)

➡️ remove_tag(...)

▶️ Member variables

▶️ Methods

➡️ __init__(...)

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ get_settings()

➡️ reset()

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ get_settings()

➡️ reset()

▶️ Member variables

▶️ Methods

➡️ __init__(...)

▶️ Member variables

▶️ Methods

➡️ __init__(...)

➡️ solve()

▶️ Member variables

▶️ Static Methods

➡️ set_log_level(...)

➡️ set_library_log_level(...)

▶️ ToolError

▶️ MaxToolErrorIter

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)

➡️ init(...)