---
id: "prereq-llm-context-windows"
type: "prereq"
source_timestamps: ["§ What Separates Leaders from Laggards", "§ A New Differentiator"]
tags: ["technical-literacy"]
related: ["entity-nathan-mapp", "action-codify-into-markdown"]
reason: "Necessary to understand the practical mechanism by which codified judgment is actually fed to the AI."
sources: ["agentic"]
sourceVaultSlug: "hbr-seg-agentic"
originDay: 6
articleStem: "hbr-new-27-teach-ai-your-decisions"
sourceUrl: "https://hbr.org/2026/06/teach-your-ai-how-you-make-decisions"
sourceTitle: "Teach Your AI How You Make Decisions"
---
# Familiarity with LLM Context Windows and Files

**Prerequisite:** Basic familiarity with how modern LLMs use provided documents ("markdown files," "context files") to ground their outputs at inference time.

**Why it matters:** Necessary to understand the practical mechanism by which codified judgment is actually fed to the AI. The [[entity-nathan-mapp|Nathan Mapp]] case ([[action-codify-into-markdown]]) and the transcript-as-context tactic ([[action-use-transcripts-as-context]]) both assume the reader knows that agents reference such files in real time — the concrete plumbing of [[concept-judgment-infrastructure]].