---
id: "concept-context-scoping"
type: "concept"
source_timestamps: ["§5.4"]
tags: ["context-engineering", "model-behavior"]
related: ["concept-five-layer-hierarchy", "claim-token-efficiency", "entity-andrej-karpathy"]
definition: "Delivering different context to the same model at each stage changes the task it performs; capability is fixed, but available information determines behaviour."
sources: ["paper"]
sourceVaultSlug: "icm-paper-folder-architecture-2026Jun02"
originDay: 2
---
# Context Scoping Changes the Task

## The mechanism

ICM's performance mechanism is **context scoping**. A model receiving:

- research instructions + source material + topic brief

…behaves differently from the same model receiving:

- a script template + voice guide + research summary.

The model's capabilities do not change between stages. What changes is the *information available* when it generates output.

## Lineage

This is context engineering in practice as framed by [[entity-andrej-karpathy-d2]]: system performance depends on *what* context is delivered, *in what structure*, and *at what moment*. ICM operationalizes the framing by structuring context into separate organizational tiers — the [[concept-five-layer-hierarchy]] — rather than a monolithic prompt.

The Layer 3 / Layer 4 distinction adds a second dimension: signalling which context is a **constraint** and which is **material to transform**.

## Connection to "lost in the middle"

Scoped context keeps relevant tokens out of the degraded mid-context band identified by Liu et al. — see [[prereq-llm-context-windows]]. This is the theoretical basis for the efficiency story in [[claim-token-efficiency]].

## Validation status

Strongly supported by broader LLM research: prompting and "role" conditioning effects are well-documented, even when the underlying model is unchanged. Tooling ecosystems like LangChain, Semantic Kernel, and [[entity-autogen]] all operationalize the same assumption via chains/graphs of prompts; ICM differs in *how* it encodes the graph (filesystem vs. code), not in the underlying premise.


## Related across days
- [[concept-dialogue-structure]]
- [[synthesis-dialogue-to-context-engineering]]
- [[prereq-llm-context-windows]]
- [[concept-five-layer-hierarchy]]