4.3 KiB
Context in Composition
Strategic framework for managing context when composing multi-agent systems.
The Core Problem
Context window is your most precious resource when composing multiple agents. A focused agent is a performant agent.
The Reality:
Single agent doing everything:
├── Context explodes to 150k+ tokens
├── Performance degrades
└── Eventually fails or times out
Multi-agent composition:
├── Each agent: <40k tokens
├── Main agent: Stays lean
└── Work completes successfully
The R&D Framework
There are only two strategies for managing context in multi-agent systems:
R - Reduce
- Minimize what enters context windows
- Remove unused MCP servers (can consume 24k+ tokens)
- Shrink static CLAUDE.md files
- Use context priming instead of static loading
D - Delegate
- Move work to sub-agents' isolated contexts
- Use background agents for autonomous work
- Employ orchestrator sleep patterns
- Treat agents as deletable temporary resources
Everything else is a tactic implementing R or D.
The Four Levels of Context Mastery
Level 1: Beginner - Stop Wasting Tokens
Focus: Resource management
Key Actions:
- Remove unused MCP servers (reclaim 20k+ tokens)
- Minimize CLAUDE.md (<1k tokens)
- Disable autocompact buffer (reclaim 20%)
Success Metric: 85-90% context window free at startup
Move to Level 2 when: Resources cleaned but still rebuilding context for different tasks
Level 2: Intermediate - Load Selectively
Focus: Dynamic context loading
Key Actions:
- Context priming (
/primecommands vs. static files) - Sub-agent delegation for parallel work
- Composable workflows (scout-plan-build)
Success Metric: 60-75% context window free during work
Move to Level 3 when: Managing multiple agents but struggling with handoffs
Level 3: Advanced - Multi-Agent Handoff
Focus: Agent-to-agent context transfer
Key Actions:
- Context bundles (60-70% transfer in 10% tokens)
- Monitor context limits proactively
- Chain multiple agents without overflow
Success Metric: Per-agent context <60k tokens, successful handoffs
Move to Level 4 when: Need agents working autonomously while you do other work
Level 4: Agentic - Out-of-Loop Systems
Focus: Fleet orchestration
Key Actions:
- Background agents (
/backgroundcommand) - Dedicated agent environments
- Orchestrator sleep patterns
- Zero-touch execution
Success Metric: Agents ship work end-to-end without intervention
When Context Becomes a Composition Issue
Trigger 1: Single Agent Exceeds 150k Tokens → Delegate to sub-agents with isolated contexts
Trigger 2: Agent Reading >20 Files → Use scout agents to identify relevant subset first
Trigger 3: /context Shows >80% Used
→ Start fresh agent, use context bundles for handoff
Trigger 4: Performance Degrading Mid-Workflow → Split workflow across multiple focused agents
Trigger 5: Same Analysis Repeated Multiple Times → Context overflow forcing re-reads; delegate earlier
Composition Patterns by Level
Beginner: Single agent, minimal static context
Intermediate: Main agent + sub-agents for parallel work
Advanced: Agent chains with context bundles for handoff
Agentic: Orchestrator + fleet of specialized agents
Key Principles
- Focused agents perform better - Single purpose, minimal context
- Agents are deletable - Free context by removing completed agents
- 200k is plenty - Context explosions are design problems, not capacity problems
- Orchestrators must sleep - Don't observe all sub-agent work
- Context bundles over full replay - 70% context in 10% tokens
Implementation Details
For practical patterns, see:
- Multi-Agent Context Isolation - Parallel execution, context bundling
- Orchestrator Pattern - Sleep patterns, fleet management
- Decision Framework - When to use each component
Source Attribution
Primary: Elite Context Engineering, Claude 2.0 transcripts Supporting: One Agent to Rule Them All, Sub-Agents documentation
Remember: Context is the first pillar of the Core 4. Master context strategy, and you can scale infinitely with focused agents.