Struct RlmExecutor 

pub struct RlmExecutor { /* private fields */ }

LLM-powered RLM executor

This is the main entry point for RLM processing. It:

1. Loads context into a REPL environment
2. Lets the LLM write analysis code
3. Executes the code and provides llm_query() for semantic sub-calls
4. Iterates until the LLM returns a FINAL answer

Implementations

impl RlmExecutor

pub fn new(context: String, provider: Arc<dyn Provider>, model: String) -> Self

Create a new RLM executor

pub fn with_max_iterations(self, max: usize) -> Self

Set maximum iterations

pub fn with_verbose(self, verbose: bool) -> Self

Enable or disable verbose mode

When verbose is true, the context summary will be displayed at the start of analysis to help users understand what’s being analyzed.

pub async fn analyze(&mut self, query: &str) -> Result<RlmAnalysisResult>

Execute RLM analysis with the given query