Validate info_level after initialization

Create LLM client

Send message to Anthropic API. :param system_prompt: System prompt string. :param messages_history: Message history list.

Encode audio file to base64 and determine format.

r""" Extract the content of the last \boxed{...} occurrence in the given text. Supports: - Arbitrary levels of nested braces

Format response for logging

Get audio duration in seconds. Tries to use wave (for .wav), then falls back to mutagen (for mp3, etc).

Determine audio file extension

Determine video file extension

Load document titles from long_context_sampled_*.json files.

Recreate the client with a fresh connection pool. This is useful when connection errors occur, as the existing connection poo

Send message event

Synchronous wrapper for image analysis. Note: Internally uses asyncio, suitable for scripts and non-async contexts.

Extract audio metadata without transcription. Use this tool to get audio file information: - Duration in seconds - Sample rate (Hz)

Answer the question based on the given audio information. Args: audio_path_or_url: The path of the audio file locally or its URL. Pa

Answer questions about audio content with confidence scoring. This tool: 1. Analyzes the audio to understand content 2. Answers the

Fast single-pass audio transcription without verification. Use this tool when: - Speed is more important than perfect accuracy - Aud

Transcribe audio file to text and return the transcription. Args: audio_path_or_url: The path of the audio file locally or its URL. P

Transcribe audio file to text and return the transcription. Args: audio_path_or_url: The path of the audio file locally or its URL. P

Advanced audio understanding with multi-turn verification and confidence scoring. This tool provides: - High-quality transcription with

Get available tokens for new content

Calculate average accuracy across individual runs

Call LLM with retry logic.

Clean up temporary file if needed. Args: file_path: Path to file is_temp: Whether file is temporary and should be delete

Clear the allowed db path restriction.

Calculate completion rate percentage

Compress content using LLM. Args: content: Content to compress level: Compression level (1=light, 2=

Compress memory if needed to fit incoming content. Args: incoming_tokens: Tokens of incoming content

Same as usual converter, but removes Javascript links and escapes URIs.

Same as usual, but be sure to start with a new line

Same as usual converter, but removes data URIs

Convert various types of resources (doc, ppt, pdf, excel, csv, zip file etc.) described by an file: or data: URI to markdown. Args: u

Count words in text, handling both Chinese and English content. For Chinese text: counts Chinese characters (each character = 1 word)

Create a linux sandbox. Args: timeout: Time in seconds before the sandbox is automatically shutdown. The default is 600 seconds. Ret

Get current total tokens in memory

Download a file from the internet to the `/home/user` dir of the remote python interpreter. Args: sandbox_id: The id of the sandbox to ru

Download a file from the sandbox to local system. Files in sandbox cannot be processed by tools from other servers - only local files and internet URL

Lazy load tiktoken encoding

Check if current message_history + summary_prompt will exceed context If it will exceed, remove the last assistant-user pair and retu

Args: batch_data: List of dicts, each containing: - case_id: str - result_text: str

Returns: (context_size, model, result_dict)

Args: Returns: (context_size, model, current_run_scores, merged_scores)

Public method to extract video metadata. Args: video_path: Path to video file Returns:

Extract tool call information from LLM response

Format processed chunks into a readable string. Args: query: The original search query chunks: List

Format token usage statistics, return summary_lines for format_final_summary and log string

Generate a summary report of all evaluation results.

Get concatenated context from top relevant chunks. Args: query: Search query max_tokens: Maximum app

Get information about a file without reading its full content. Useful for understanding file structure before reading specific parts. Arg

Get memory statistics

Perform google searches via Serper API and retrieve rich results. It is able to retrieve organic search results, people also ask, related searches

Tool to perform web searches via Serper API and retrieve rich results. It is able to retrieve organic search results, people also ask, r

Perform google searches via Serper API and retrieve rich results. It is able to retrieve organic search results, people also ask, related searches

Handle max turns reached summary prompt