Skip to content

API Chat

Jump to bottom
github-actions[bot] edited this page Oct 3, 2025 · 7 revisions

answer_rocket.chat

Classes

Chat

Methods

__init__(self, gql_client: GraphQlClient, config: [ClientConfig](API-Types#clientconfig))
ask_question(self, copilot_id: str, question: str, thread_id: str, skip_report_cache: bool, dry_run_type: str, model_overrides: dict, indicated_skills: list[str], history: list[dict], question_type: [QuestionType](API-Types#questiontype), thread_type: [ThreadType](API-Types#threadtype)) -> MaxChatEntry

Calls the Max chat pipeline to answer a natural language question and receive analysis and insights in response.

Parameters:

  • copilot_id (str): The ID of the copilot to run the question against.
  • question (str): The natural language question to ask the engine.
  • thread_id (str, optional): ID of the thread/conversation to run the question on. The question and answer will be added to the bottom of the thread.
  • skip_report_cache (bool, optional): Should the report cache be skipped for this question? Defaults to False.
  • dry_run_type (str, optional): If provided, run a dry run at the specified level: 'SKIP_SKILL_EXEC', 'SKIP_SKILL_NLG'.
  • model_overrides (dict, optional): A dictionary of model types to model names to override the LLM model used. Model type options are 'CHAT', 'EMBEDDINGS', 'NARRATIVE'.
  • indicated_skills (list[str], optional): A list of skill names that the copilot will be limited to choosing from. If only 1 skill is provided the copilot will be guaranteed to execute that skill.
  • history (list[dict], optional): A list of messages to be used as the conversation history for the question.
  • question_type (QuestionType, optional): The type of question being asked. This is used to categorize the question and can determine how the UI chooses to display it.
  • thread_type (ThreadType, optional): The type of thread being created. This is used to categorize the thread and can determine how the UI chooses to display it.

Returns:

MaxChatEntry - The ChatEntry response object associated with the answer from the pipeline.

add_feedback(self, entry_id: str, feedback_type: [FeedbackType](API-Types#feedbacktype), feedback_text: str) -> bool

This adds feedback to a chat entry.

Parameters:

  • entry_id (str): The id of the chat entry.
  • feedback_type (FeedbackType): The type of feedback to add.
  • feedback_text (str, optional): The text of the feedback.

Returns:

bool - True if the feedback was added successfully, False otherwise.

get_threads(self, copilot_id: str, start_date: datetime, end_date: datetime) -> list[MaxChatThread]

Fetches all threads for a given copilot and date range.

Parameters:

  • copilot_id (str): The ID of the copilot to fetch threads for.
  • start_date (datetime, optional): The start date of the range to fetch threads for.
  • end_date (datetime, optional): The end date of the range to fetch threads for.

Returns:

list[MaxChatThread] - A list of ChatThread objects.

get_entries(self, thread_id: str, offset: int, limit: int) -> list[MaxChatEntry]

Fetches all entries for a given thread.

Parameters:

  • thread_id (str): The ID of the thread to fetch entries for.
  • offset (int, optional): The offset to start fetching entries from.
  • limit (int, optional): The maximum number of entries to fetch.

Returns:

list[MaxChatEntry] - A list of ChatEntry objects.

evaluate_entry(self, entry_id: str, evals: list[str]) -> EvaluateChatQuestionResponse

Runs and fetches the inputted evaluations for a given entry.

Parameters:

  • entry_id (str): The ID of the entry to fetch evaluation for.
  • evals (list[str]): A list of strings containing the evaluations to run on the entry.

Returns:

EvaluateChatQuestionResponse - The evaluation response object containing evaluation results.

share_chat_thread(self, original_thread_id: str) -> SharedThread

Share a chat thread by its ID.

Parameters:

  • original_thread_id (str): The ID of the original thread to share.

Returns:

SharedThread - The shared thread object.

get_chat_entry(self, entry_id: str) -> MaxChatEntry

Retrieve a chat entry by its ID.

Parameters:

  • entry_id (str): The ID of the chat entry to retrieve.

Returns:

MaxChatEntry - The chat entry object.

get_chat_thread(self, thread_id: str) -> MaxChatThread

Retrieve a chat thread by its ID.

Parameters:

  • thread_id (str): The ID of the chat thread to retrieve.

Returns:

MaxChatThread - The chat thread object.

create_new_thread(self, copilot_id: str, thread_type: [ThreadType](API-Types#threadtype)) -> MaxChatThread

Create a new chat thread for the specified agent.

Parameters:

  • copilot_id (str): The ID of the agent to create the thread for.
  • thread_type (ThreadType, optional): The type of thread to create. Defaults to CHAT. For most purposes CHAT is the only type needed.

Returns:

MaxChatThread - The newly created chat thread object.

queue_chat_question(self, question: str, thread_id: str, skip_cache: bool, model_overrides: dict, indicated_skills: list[str], history: list[dict]) -> MaxChatEntry

This queues up a question for processing. Unlike ask_question, this will not wait for the processing to complete. It will immediately return a shell entry with an id you can use to query for the results.

Parameters:

  • question (str): The text of the user's question.
  • thread_id (str): ID of the thread the question is being sent to.
  • skip_cache (bool, optional): Set to true to force a fresh run of the question, ignoring any existing skill result caches. Defaults to False.
  • model_overrides (dict, optional): A dictionary of model types to model names to override the LLM model used. Model type options are 'CHAT', 'EMBEDDINGS', 'NARRATIVE'.
  • indicated_skills (list[str], optional): A list of skill names that the copilot will be limited to choosing from. If only 1 skill is provided the copilot will be guaranteed to execute that skill.
  • history (list[dict], optional): A list of messages to be used as the conversation history for the question.

Returns:

MaxChatEntry - A shell entry with an id you can use to query for the results.

cancel_chat_question(self, entry_id: str) -> MaxChatEntry

This deletes the entry from its thread and attempts to abandon the question's processing if it is still ongoing.

Parameters:

  • entry_id (str): The id of the chat entry.

Returns:

MaxChatEntry - The deleted entry.

get_user(self, user_id: str) -> MaxChatUser

This fetches a user by their ID.

Parameters:

  • user_id (str): The id of the user.

Returns:

MaxChatUser - A MaxChatUser object.

get_all_chat_entries(self, offset, limit, filters) -> list[MaxChatEntry]

Fetches all chat entries with optional filters.

Parameters:

  • offset (int, optional): The offset to start fetching entries from. Defaults to 0.
  • limit (int, optional): The maximum number of entries to fetch. Defaults to 100.
  • filters (dict, optional): A dictionary of filters to apply to the query. Supports all filtering available in the query browser.

Returns:

list[MaxChatEntry] - A list of ChatEntry objects.

get_skill_memory_payload(self, chat_entry_id: str) -> dict

Fetches the skill memory payload for a given chat entry.

Parameters:

  • chat_entry_id (str): The id of the chat entry.

Returns:

dict - The skill memory payload for the given chat entry.

set_skill_memory_payload(self, skill_memory_payload: dict, chat_entry_id: str) -> bool

Sets the skill memory payload for a given chat entry.

Parameters:

  • skill_memory_payload (dict): The skill memory payload to set -- must be JSON serializable.
  • chat_entry_id (str, optional): The id of the chat entry.

Returns:

bool - True if the skill memory payload was set successfully, False otherwise.

get_dataframes_for_entry(self, entry_id: str) -> [pd.DataFrame]

This fetches the dataframes (with metadata) for a given chat entry.

Parameters:

  • entry_id (str): The answer entry to fetch dataframes for.

Returns:

list[pd.DataFrame] - A list of dataframes and metadata for the given chat entry.

get_chat_artifact(self, chat_artifact_id: UUID) -> Optional[ChatArtifact]

Retrieve a chat artifact by its ID.

This method queries the backend for a chat artifact using the given unique identifier. If the artifact is found, it is returned as a ChatArtifact object. If not found or if an error occurs during the query, None is returned.

Parameters:

  • chat_artifact_id (UUID): The unique identifier of the chat artifact to retrieve.

Returns:

ChatArtifact or None - The chat artifact object if found, or None if not found or if an error occurs.

get_chat_artifacts(self, search_input: Optional[ChatArtifactSearchInput], paging: Optional[PagingInput]) -> PagedChatArtifacts

Retrieve paged chat artifacts based on optional search and paging criteria.

If no search_input or paging is provided, default values will be used.

Parameters:

  • search_input (ChatArtifactSearchInput, optional): An object specifying the search criteria for chat artifacts. If None, no filters are applied on name or misc_info.
  • paging (PagingInput, optional): An object specifying pagination details such as page number and page size. If None, defaults to page 1 with a page size of 100.

Returns:

PagedChatArtifacts - A paged collection of chat artifacts. Returns an empty PagedChatArtifacts instance if an error occurs during retrieval.

create_chat_artifact(self, chat_artifact: ChatArtifact) -> MaxMutationResponse

Submits a GraphQL mutation to create a new chat artifact.

Parameters:

  • chat_artifact (ChatArtifact): The chat artifact object containing the data to be created.

Returns:

MaxMutationResponse - The response object containing the result of the mutation, specifically the created chat artifact.

delete_chat_artifact(self, chat_artifact_id: uuid) -> MaxMutationResponse

Submits a GraphQL mutation to delete an existing chat artifact by its ID.

Parameters:

  • chat_artifact_id (uuid): The UUID of the chat artifact to delete.

Returns:

MaxMutationResponse - The response object containing the result of the deletion mutation.

Clone this wiki locally