rlm

Externalize AI agent state and orchestrate recursive subproblem processing across files.

1|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/chrisliu298/dotfiles --skill rlm-chrisliu298
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rlm
Source: https://github.com/chrisliu298/dotfiles/tree/main/agents/extensions/skills/rlm
Command: npx skills add https://github.com/chrisliu298/dotfiles --skill rlm-chrisliu298

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Handles data-scale AI tasks by externalizing large state to files and guiding recursive decomposition, avoiding context overload and ensuring coverage across many items.

Core Features & Use Cases

  • Externalize state and generate subproblems from manifests for batched processing
  • Programmatic delegation to distribute work across subagents and batches
  • File-based output artifacts with final summaries for audit and reproducibility

Quick Start

Clone the rlm skill into your agent's skills directory and run the controller to start externalizing state, generating sub-problems, and assembling deliverables in files.

Frequently Asked Questions about rlm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I process large, multi-file datasets with AI without overflowing the chat context?

Externalizing state to files and applying recursive decomposition handles large, multi-file datasets, preventing chat context overflow while ensuring complete coverage across many items.

What is recursive problem decomposition for agent orchestration?

Recursive problem decomposition generates subproblems from manifests for batched processing, enabling programmatic delegation to distribute computational work across subagents and data batches.

How do I start batched processing with externalized state for dense data tasks?

Clone the skill into your agent's directory and run the controller to start externalizing state, generating sub-problems, and assembling deliverables in file-based output artifacts.

Can I use agent orchestration for cross-reference reasoning across many files?

Yes, agent orchestration applies to tasks involving many files, dense data, or complex cross-reference reasoning where a chat context would overflow and per-item processing is needed.

How do I ensure reproducibility and auditability when delegating work to subagents?

Generate file-based output artifacts with final summaries to ensure reproducibility and auditability, recording assembled deliverables from distributed subagent processing.