---

name: rlm-reasoning description: Apply RLM (Recursive Language Model) reasoning pattern for large context processing. Use when analyzing large files, codebases, or documents. Triggers on "RLM pattern", "large file analysis", "selective reading", "PROBE EXTRACT CONFIRM", or when dealing with files over 1000 lines. allowed-tools:

• Read
• Grep
• Glob
• Task
• Bash

---

RLM (Recursive Language Model) Reasoning Pattern

Based on Zhang et al., MIT CSAIL (arXiv:2512.24601)

Core Principle: Context Outside, Not Inside

Never load entire large files into context. Instead, keep data "outside" and access selectively.

4-Step Reasoning Pattern

1. PROBE (Explore)

Search for relevant areas using keywords before reading:

CopyGrep for keywords to find relevant lines
Glob to find related files by pattern

Example:

CopyGrep(pattern="authentication", path="src/")
Glob(pattern="**/auth*.py")

2. EXTRACT (Load Selectively)

Only read the specific sections you need:

CopyRead with offset/limit for specific line ranges
Load only chunks that contain matches

Example:

CopyRead(file_path="src/auth/login.py", offset=50, limit=100)

3. CONFIRM (Verify)

Before answering, verify information against the source:

CopyRe-check that found information matches the question
Validate references actually exist
Cross-reference multiple sources if needed

4. ANSWER (Respond)

Provide verified response based on confirmed evidence.

When to Apply This Pattern

Anti-Patterns (Avoid)

Copy# BAD: Loading entire file
Read("large_file.py")  # Don't do this!

# GOOD: Search first, then selective read
Grep("function_name", "src/")
Read("src/file.py", offset=150, limit=50)

Sub-Agent Delegation

For complex multi-step analysis, use Task tool:

CopyTask(
    subagent_type="Explore",
    prompt="Find all authentication-related code and summarize the flow"
)

Verification Checklist

Before providing an answer:

• Did I search before reading?
• Did I read only necessary sections?
• Did I verify references exist?
• Is my answer based on confirmed evidence?

Additional Resources

• For Python implementation details, see rlm_agent.py
• For detailed API reference, see REFERENCE.md

Quick Commands

Run the RLM agent directly:

Copypython ~/.claude/skills/rlm-reasoning/scripts/rlm_query.py "your question" /path/to/file