> ## Documentation Index > Fetch the complete documentation index at: https://praison.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # extras • Rust AI Agent SDK > Extra parity types and functions # extras Rust AI Agent SDK Extra parity types and functions This module implements additional Python SDK features for full parity: * Deep Research types (DeepResearchResponse, ReasoningStep, etc.) * RAG types (RAGCitation, RetrievalPolicy, etc.) * Guardrail types (LLMGuardrail) * Handoff errors * App protocols * Embedding functions * Module re-exports ## Import ```rust theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} use praisonai::parity::extras::*; ``` ## Classes Represents a citation in the research report Represents a reasoning step in the research process Represents a web search call made during research Represents a code execution step during research Represents a file search call (Gemini-specific) Complete response from a Deep Research query LLM-based guardrail for content validation Agent application configuration Security policy for sandbox execution Output from a reflection step Embedding result Embedding usage statistics Trace context for tracking operations Agent application protocol Agent OS protocol (alias) Database adapter trait Observability collector trait Supported Deep Research providers Retrieval policy for RAG Base error type for handoff operations ## Functions Synchronous embedding function Synchronous embedding function (alias) Synchronous embeddings function (batch) Async embedding function Async embedding function (alias) Async embeddings function (batch) Get sync display callbacks Get async display callbacks Get error logs Resolve guardrail policies Get or create trace context Track workflow execution Load a plugin from a path