Loading patterns…
Context Orchestration
Patterns for selecting, routing, sizing, and combining context
In 30 seconds
- What
- Decides which information reaches each model or agent, in what form, and when, by selecting and routing across memory, retrieval, tools, and specialists.
- When to use
- Multiple context sources compete for limited model window, different agents need different evidence, or context quality and freshness vary by source.
- Watch out
- Concatenating every available source without prioritization wastes tokens, introduces noise, and can trigger conflicting reasoning or privacy leaks.
Ask the AI expert about these patterns
Opens the assistant with your question prefilled. You review it before sending.
Overview
Context orchestration patterns decide which information should reach each model or agent, in what form, and at what time. They cover source fusion, task-aware routing, and adaptive context sizing across systems that combine memory, retrieval, tools, and multiple specialists.
Practical Applications & Use Cases
Multi-source assistants
Merge retrieved documents, session memory, and live tool results with clear provenance.
Specialist agents
Route only the context each agent needs while preserving handoff continuity.
Cost-sensitive applications
Adapt context depth to task complexity instead of filling the window by default.
Why This Matters
More context is not always better. Deliberate orchestration reduces distraction, leakage, cost, and conflicting evidence while preserving the information needed for a task.
Implementation Guide
When to Use
- Several context sources compete for a limited model window
- Different agents or tasks need different evidence
- Context quality, privacy, or freshness varies by source
Best Practices
- Rank context by task relevance, authority, freshness, and sensitivity
- Keep provenance and access policy attached through transformations
- Evaluate retrieval and answer quality together
Common Pitfalls
- Concatenating every available source without prioritization
- Dropping provenance during summarization or handoff
- Sharing sensitive context with agents that do not need it
Available Techniques
Multi-Source Context Fusion(MSCF)
Intelligently combines contextual information from multiple sources with quality weighting and conflict resolution
Intelligent Context Routing(ICR)
Dynamic routing of contextual information to the most appropriate processing components based on capability matching
Dynamic Context Assembly(DCA)
Intelligent, real-time composition of context from multiple sources based on query analysis and relevance scoring
Adaptive Context Depth(ACD)
Dynamically adjust memory depth and context complexity based on task requirements in multi-agent agentic AI systems
Patterns Pack
Take the whole catalog with you: MCP server, editor rules and skills, and data.
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.
From the engineer behind this catalog
Get your agent architecture reviewed
This page documents one pattern. Your system runs dozens, and most failures live in how they fit together. Have the whole design reviewed against the 288 patterns in this catalog: architecture, reliability, evaluation and cost, every finding mapped to the pattern that fixes it.
€750 instead of €1,500, one week, written report and walkthrough call, until 30 September