Rlm

Split a file into overlapping chunks suitable for parallel fanout processing and emit a manifest describing each chunk

Dispatch the same query to multiple subagents in parallel across chunks in a manifest and aggregate results with source provenance

Detect requests for recursive decomposition and large-scale operations that benefit from RLM processing

Prepare source content for RLM processing by discovering files, chunking each one, and writing a unified searchable manifest

Parallel fan-out processing - spawn multiple sub-agents for chunked context processing

Spawn sub-agent to process focused context and return structured result

Run the full Recursive Language Model pipeline — prep, fan out across chunks, and recursively synthesize until results fit one context window