Recursive Language Model (RLM)

The RLM Loop

Phase 1: Index & Filter

# Find candidate files
grep -rl "pattern" src/ --include="*.ts"
find . -name "*.py" -newer last_check

Phase 2: Parallel Map

• Launch 3-5+ agents in parallel for broad tasks
• Give each agent ONE specific file or chunk
• Each agent returns a structured summary

Agent 1: "Read src/api/routes.ts. List all endpoints with their auth decorators."
Agent 2: "Read src/api/users.ts. List all endpoints with their auth decorators."
...

Phase 3: Reduce & Synthesize

Critical Rules

1. NEVER read more than 3-5 files into your main context
2. ALWAYS use parallel agents when file count > 5
3. Write Python scripts for state tracking across 50+ files — let the script scan and summarize
4. If parallel agents are unavailable, fall back to iterative Python scripting

Example: "Find all API endpoints, check for Auth"

1. grep -l "@Controller" src/**/*.ts

   → 20 files
2. Spawn 20 agents, each extracts endpoints + auth status
3. Collect outputs, compile table, identify missing auth

Output Format

• Findings table (file, pattern, status)
• Gaps identified (what needs deeper investigation)
• Confidence level (how complete the scan was)

Skill Boundaries