Get all relationships for a specific entity in the knowledge graph. This tool explores the knowledge graph to find how a specific entity

Get relationships for an entity. Args: input_data: Entity relationship parameters Returns: Entity relationships

Get timeline of facts for an entity using Graphiti. Args: entity_name: Name of the entity start_date

Get timeline of facts for an entity. Args: input_data: Timeline query parameters Returns: Timeline of facts

Returns the features data: template, mandatory fields The features are related with the form buttons

Returns the features data: template, mandatory fields The features are related with the from-prompt form buttons

Returns the fields data: description, type, length

Returns the fields data for ideation from question (prompt)

Retrieves and decodes the content of a file from its base64 representation. This function validates the file using `validate_file` and then decod

Fetch games for a specific date

Get basic statistics about the knowledge graph. Returns: Graph statistics

Returns the response from the index

Get ingestion-specific LLM model (can be faster/cheaper than main model). Returns: Configured model for ingestion tasks

Get model for ingestion tasks (uses same model as main LLM). Returns: Configured model for ingestion tasks

Get model for ingestion tasks (uses same model as main LLM). Returns: Configured model for ingestion tasks

Get model for ingestion tasks (uses same model as main LLM). Returns: Configured model for ingestion tasks

Returns the item in the database

Returns a list of video URLs Args: item_type (str): The type of item to get the URLs for. E.g. "video" or

Fetches the most recent video from a public YouTube playlist.

Returns the items in the database

Returns the items in the database

Extract the live chat ID from the provided metadata. This function attempts to retrieve the active live chat ID from the 'liveStreamingDetail

Get LLM model configuration for OpenAI. Returns: Configured OpenAI model

Get LLM model configuration for OpenAI. Returns: Configured OpenAI model

Get the LLM provider name.

Initialize and return the primary agent with all MCP servers running. This function sets up an AsyncExitStack and starts all MCP servers,

Extract metadata from a document file. Args: file_path: Path to the document file Returns:

Get metadata for a PDF file including PDF-specific properties. Args: file_path: Path to the PDF file

Configure and return the LLM model to use.

Returns the model options for the LLM call

Get parlay prediction for games on a specific date.

Get URLs from Pydantic AI docs sitemap.

Fetches all URLs from the Pydantic AI documentation. Uses the sitemap (https://ai.pydantic.dev/sitemap.xml) to get these URLs. Retur

Fetches all URLs from the Pydantic AI documentation. Uses the sitemap (https://ai.pydantic.dev/sitemap.xml) to get these URLs. Retur

Fetches all URLs from the Pydantic AI documentation. Uses the sitemap (https://ai.pydantic.dev/sitemap.xml) to get these URLs. Retur

Fetches all URLs from the Pydantic AI documentation. Uses the sitemap (https://ai.pydantic.dev/sitemap.xml) to get these URLs. Retur

Returns the question label based on the tab

Returns the reference files to be used

Get entities related to a given entity using Graphiti search. Args: entity_name: Name of the entity

Retrieve all messages for a given session.

Get current standings.

Retrieves stream metadata for a given YouTube video ID using the YouTube API. This function takes a YouTube video ID, calls the YouTube API

Get suggestions from the AI

Extract title and summary using GPT-4.

Extracts the title and summary from a text chunk using a buzz_intern_agent. This function uses buzz_intern_agent to process the input chunk and e

Initialize and return a Twitter API client using OAuth 1.0a.

Returns the unified flag. Returns: bool: True if the model or provider is not allowed to have a system

Returns the use embeddings button config

Returns the user input text or file

Groups unwritten chat messages by session ID, live chat ID, and retry count. This function takes a list of unwritten chat messages and group

Convert a FastAPI headers object to a dictionary.

Perform both vector and keyword search for comprehensive results. This tool combines semantic similarity search with keyword matching

Perform hybrid search (vector + keyword). Args: embedding: Query embedding vector query_text: Query text for keyword sea

Import data into the database

Ingest all documents from the documents folder. Args: progress_callback: Optional callback for progress updates

Ingest all documents from the documents folder. Args: progress_callback: Optional callback for progress updates

Ingest all documents from the documents folder. Args: progress_callback: Optional callback for progress updates

Ingest all documents from the documents folder. Args: progress_callback: Optional callback for progress updates

Initialize database connections.

Initialize database connections.

Initialize database connections.

Initialize the server connection.

Initialize the server connection.

Initialize database connections.

Load environment variables, resolve provider-specific configuration, and return the client and model. Returns: tuple[AsyncOpenAI

Initialize database connection pool.

Initialize database connection pool.

Initialize database connection pool.

Initialize cross-encoder model for re-ranking.

Insert a processed chunk into Supabase.

Send an interaction request to Voiceflow Dialog API. NOTE: Right now this only supports text responses, but in the future it will support

Run the agent with streaming text for the user_input prompt, while maintaining the entire conversation in `st.session_state.messages`.

Determine if this is a continuing conversation or a new one. Args: messages: List of messages in the request Returns: T