rlm-mode

Detect large-scale code requests and route them to RLM processing commands.

175|26|Updated Aug 14, 2025
One-click install
npx skills add https://github.com/jmagly/aiwg --skill rlm-mode-jmagly
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rlm-mode
Source: https://github.com/jmagly/aiwg/tree/main/agentic/code/addons/rlm/skills/rlm-mode
Command: npx skills add https://github.com/jmagly/aiwg --skill rlm-mode-jmagly

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents context window overflow and performance degradation when users request operations on entire codebases or large numbers of files, by intelligently routing requests to specialized RLM (Recursive Language Model) processing commands.

Core Features & Use Cases

  • Detects Large-Scale Requests: Identifies phrases indicating operations across many files (e.g., "analyze all files", "search the entire codebase").
  • Routes to RLM Commands: Suggests and formats appropriate RLM commands (rlm-query, rlm-batch, rlm-summarize) based on the user's intent.
  • Handles Ambiguity: Prompts the user for clarification when the scope of the request is unclear.
  • Use Case: A user asks to "find all TODO comments across the entire project." This Skill recognizes the scope, suggests using /rlm-query "Extract TODO comments" --path . --pattern "**/*.{js,ts,jsx,tsx}", and explains the benefits of RLM processing.

Quick Start

Use the rlm-mode skill to find all TODO comments across the entire project.

Frequently Asked Questions about rlm-mode

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

FAQPage Schema
How do I perform large-scale codebase analysis without hitting context window limits?

Recursive processing handles large-scale codebase analysis by decomposing tasks into manageable chunks, preventing context window overflow and performance degradation during batch operations across extensive file patterns.

What is the best way to search the entire codebase for specific comments?

The best way to search the entire codebase for specific comments is using specialized query commands with directory scopes and file pattern parameters. This approach uses hierarchical summarization to manage large-scale operations efficiently.

Can I process batch operations across multiple files with specific directory scopes?

Yes, you can process batch operations across multiple files by specifying directory scopes and depth parameters. The system routes these large-scale requests to specialized commands that execute recursive processing for precise file pattern matching.

Why does my codebase analysis fail with context window overflow on large projects?

Codebase analysis fails with context window overflow on large projects because the volume of files exceeds processing capacity. Recursive Language Model processing solves this by breaking down large-scale operations into hierarchical summarization and batch chunks.

When do I need recursive processing for code operations?

You need recursive processing for code operations when handling requests across entire codebases or large numbers of files. It detects large-scale operations and routes them to specialized commands to prevent context window overflow and performance degradation.