🧠 Advanced Patterns

RLM, Ralph-RLM, GasTown, and multi-agent orchestration patterns

RLM (Recursive Language Model)

Hierarchical task decomposition with persistent context

Production Ready

The RLM pattern breaks complex tasks into hierarchical subtasks, with each level storing findings in REMBR. Sub-agents search context, validate results, and synthesize insights.

How It Works

  • •L0 agent receives complex task
  • •Decomposes into L1 subtasks
  • •Each L1 stores findings in REMBR
  • •L0 searches context, synthesizes

REMBR Tools Used

  • •create_context - Task workspaces
  • •store_memory - Save findings
  • •search_context - Retrieve results
  • •create_snapshot - State handoff

Ralph-RLM (Acceptance-Driven RLM)

Loops until explicit criteria are met with stuck detection

Production Ready

The Ralph-RLM pattern combines RLM decomposition with validation loops. Tasks iterate until all acceptance criteria pass, with automatic stuck detection and plan regeneration.

How It Works

  • •Store acceptance criteria in REMBR
  • •Execute RLM decomposition
  • •Validate against criteria
  • •Loop until all criteria met or stuck

Key Features

  • •Explicit acceptance criteria
  • •Automatic stuck detection
  • •Plan regeneration on stall
  • •Progress tracking in REMBR

GasTown (Multi-Agent Collaboration)

Parallel agents with shared context and conflict resolution

Experimental

The GasTown pattern enables parallel agents to collaborate on tasks using REMBR as a shared knowledge base with conflict detection and resolution.

How It Works

  • •Multiple agents share REMBR context
  • •Each agent stores findings
  • •Contradiction detection via REMBR
  • •Coordinator resolves conflicts

REMBR Tools Used

  • •detect_contradictions
  • •get_memory_graph
  • •infer_causality
  • •compare_snapshots

🔗 Combining Patterns

Mix and match for complex workflows

RLM + Ralph-RLM
Hierarchical decomposition with validation loops at each level
RLM + GasTown
Parallel sub-agents collaborate on L1 tasks, synthesize at L0
Ralph-RLM + GasTown
Team validates against acceptance criteria with conflict resolution