Answer a question about user conversation history using RAG Args: question: The question to answer t

Send a message to the agent Args: message: User message use_tools: Whether to enable tool calling

Choose an action using epsilon-greedy strategy.

Clear conversation history

Execute code in a sandboxed environment with multi-language support. Args: code: Code to execute lan

Count the number of tokens in a text string.

Factory function to create appropriate memory manager Args: user_id: User identifier mode: Memory mode (defaults to conf

Evaluate an agent's response against the test case criteria. Args: test_case: The test case being evaluated

Evaluate a single test case Args: test_id: The test case ID to evaluate Returns:

Get API key for specified provider (or configured provider if not specified)

Get OpenAI client configuration

Get company information and business details. Args: symbol: Stock ticker symbol Returns: TextContent with c

Retrieve historical stock data. Args: symbol: Stock ticker symbol start: Start date (YYYY-MM-DD) end: End date (

Get configuration for a specific model

Get the configured provider

Increment the conversation counter

Initialize browser session.

Process a user query using the ReAct pattern. Args: user_query: The user's question stream: Whether

Request LLM approval for a dangerous operation. Args: operation: The operation name details: Details

Answer a query given extracted text context

Get element using stable selectors with fallback. Args: step: The step containing selector information

Create a hashable representation of the game state. This is crucial for tabular Q-learning.

Add a conversation turn Args: session_id: Session identifier user_message: User's message

Calculate statistics for a list of rewards

Calculate win probability of model_a against model_b based on current ratings. Args: model_a: First model identi

Main chat interface with streaming support

Chunk a single conversation into segments Args: conversation_id: Unique conversation identifier test

Chunk a document with optional contextual enhancement. Educational Note: This is the core innovation - each chunk ge

Clear the chain results

Clear all documents.

Perform contextual search using specified method. Args: query: Search query method: Search method -

Cleanup MCP manager (no persistent connections to close) Since we spawn fresh connections for each tool call, there'

Download file from URL to temporary location. Args: url: URL to download from timeout: Request timeout in seconds

Evaluate a single test case with dual memory system Args: test_id: Test case ID Returns:

Execute the research task Args: max_iterations: Maximum number of tool calls Returns:

Get base URL for the configured provider

Get or create the global PubChem client.

Retrieve detailed chemical properties for a PubChem compound. Args: cid: PubChem Compound ID properties: List of propert

Get statistics about the knowledge graph.

Get GraphRAG configuration from environment.

Return the hidden game rules (for debugging/analysis).

Get RAPTOR configuration from environment.

Get current rating for a model, initializing if necessary.

Get chunking statistics

Get comprehensive statistics

Get statistics about the RAPTOR tree.

Add event to Google Calendar. Args: summary: Event title start_time: Start time (ISO 8601 format or

Handle an incoming event and generate a response Args: event: The event to handle max_iterations: Ma

Check if the given string is a URL. Args: path_or_url: String to check Returns: True if the string is a URL

Terminate the shell session

Load test cases from the evaluation framework

Make a rate-limited request to PubChem API with retry for async operations.

Play one episode of the game.

Print formatted result

Print a formatted section header

Process conversation history with both contextual chunking and memory cards. Args: chunks: Basic conversation ch

Read a text file with size limits. Args: file_path: Path to the file encoding: File encoding

Reset conversation history

Run the interactive CLI

Save notes to storage

Save JSON memory cards to storage

Save advanced JSON memory cards to storage

Save memory cards to storage

Save a workflow to persistent storage. Args: workflow: The workflow to save

Search the index for relevant chunks using retrieval pipeline Args: query: Search query top_k: Numbe

Search for documents matching the query

Send a Slack message via webhook. Args: message: Message text to send webhook_url: Optional Slack webhook URL (uses default i

Get the number of documents.

Stop the vLLM server

Convert event to dictionary for JSON serialization

Undo the last move. Returns: Dictionary with board state after undo

Validate configuration

Execute shell command in a virtual terminal. Args: command: Shell command to execute timeout: Timeou

Write data to Excel file. Args: file_path: Path to Excel file data: Dictionary of {sheet_name: [rows]} overwrite

Initialize memory manager Args: user_id: Unique identifier for the user verbose: Whether to print de

Extract PDF to text using OCR with thinking mode

Extract unique tools used from trajectory. Args: trajectory: Execution trajectory Returns:

Format file size in human-readable format.

Initialize the system prompt