Prompt Engineering

Explain Large Class — the 900-Line Walkthrough

The class everyone avoids: its actual responsibilities, its key methods, its side effects, and the paths through it — understood before anyone proposes splitting it.

Overview

Giant classes resist understanding precisely because narration fails at their scale — 900 lines of "then it does this" teaches nothing. This prompt uses the function-walkthrough mode at class scope: the contract of the class (what it is given, what it produces, what else changes), the responsibilities it actually performs — usually more than its name admits — its key methods walked with decision points and both branches, and every side effect surfaced with its timing. The honesty discipline matters at this scale: claims about why the class grew this way are inferences, labeled as such; the open questions list what reading alone cannot settle.

Workflow

State the class contract
What it is given, what it produces, what else changes — the class as one unit first.
Find the real responsibilities
The jobs it actually performs, usually more than the name admits — each mapped to its methods.
Walk the load-bearing methods
Decision points with both branches, side effects with timing — depth where the behavior lives.

Why This Works

Class-as-contract framing survives a scale where line narration collapses
Responsibility mapping is the understanding a future split actually needs
Labeled inference keeps the class's history honest — growth stories are guesses

Best for

Classes that grew for years and resist reading
Pre-refactor understanding (the step everyone skips)
Newcomers assigned to the module nobody explains

Not for

Actually decomposing the class — that's the Refactor Prompt Builder's Large Method Decomposition territory
Judging the class as over-grown — that's the Code Review Prompt Generator's maintainability focus

Use cases

Understanding the god class before proposing its split
Mapping which methods carry the real behavior
Surfacing the side effects buried in 900 lines

Customize This Resource

Opens this setup in Code Explanation Prompt. Generate to get the full understanding contract — then adjust the mode, audience, and depth.

Open in Code Explanation Prompt

Prompt Template

EXPLANATION OBJECTIVE
Walk through the 900-line OrderManager class: its responsibilities, key methods, and side effects.
Explanation mode: function walkthrough — inputs, outputs, execution flow, decisions, and side effects.
The goal is understanding, not judgment: explain what this code does and why it is built this way — and clearly separate what the code shows from what you infer.

CODE CONTEXT
System context: The class everyone avoids touching; it grew for eight years.
Code to explain:
```
[Paste the code to explain here]
```

AUDIENCE PROFILE
Audience: a mid-level developer — fluent in the language and common patterns.
- Assume: language features and common patterns are known; this codebase's conventions and domain are not.
- Skip syntax explanations entirely; spend the depth on intent, structure, and domain meaning.

EXPLANATION STRATEGY
1. State the contract first: inputs with their expected shapes, outputs with their meaning, errors with their triggers.
2. Walk the execution flow in order, pausing at every decision point to explain both branches.
3. Surface every side effect — writes, calls to other systems, mutations of shared state — and when each happens.
4. Distinguish the happy path from the edge paths; name the conditions that route execution off the happy path.
- Depth: detailed explanation — thorough coverage of behavior and structure.
- Cover every section of this contract with real content — no skipped or token sections.
- Go as deep as the code requires, but no deeper: detail serves understanding, not completeness theater.

KEY CONCEPTS
- Explain each non-obvious variable, structure, and helper the walkthrough touches — what it holds and why it exists.
- Order concepts by how load-bearing they are for understanding this code — not by where they appear in the file.

EXECUTION FLOW
- Present the flow as a numbered sequence; at each branch, state the condition in plain language and what each outcome means.
- Where the flow depends on state or configuration, say which, and what each setting changes.

DESIGN DECISIONS
- Where the implementation makes a visible choice — an early return, a special case, a particular ordering — explain what the choice achieves.
- Discuss only decisions actually visible in the code; rationale you cannot evidence belongs under ASSUMPTIONS, labeled as inference.

ASSUMPTIONS
- Separate what the code DOES (verifiable from the source) from why it might be that way (inference).
- Label every statement about intent, history, or rationale as an INFERENCE unless the code itself evidences it — a comment, a test, a telling name.
- Where behavior cannot be determined from the provided code, say "cannot be determined from this code" — never fill the gap with a plausible story.

TRADEOFFS
- Identify the tradeoffs this code embodies: what it optimizes for, and what it sacrifices for that.
- Tie each tradeoff to its visible consequence in the code.

LEARNING NOTES
- End with the contract restated in two sentences: what you give it, what you get back, and what else changes.

NON-GOALS
- Do not review code quality — no findings, no severities, no verdicts.
- Do not debug failures — explain the behavior that exists without diagnosing it.
- Do not refactor the code or propose rewrites.
- Do not generate tests.
- Do not invent requirements or unstated intentions.
- If something looks defective, note it as a question for the team — investigating or fixing it is out of scope.
- Clearly separate facts from assumptions throughout.

OUTPUT REQUIREMENTS
- Structure the explanation with headings that mirror the sections above — a reader should be able to navigate it without reading all of it.
- Anchor every claim about behavior in the code: name the function, branch, or line it comes from when precision matters.
- Calibrate every section to the stated audience — too basic wastes their time, too advanced loses them.
- End with the open questions: what could not be determined from the provided code, and who or what could answer it.

Related Resources

Explore more resources that work well with this one.

Used in Playbooks

Complete workflows that include this resource as a step.

AI Codebase Onboarding Workflow Workflow · 4 steps

Tip: Save time by exploring related resources and tools that integrate with this workflow.

Explore all resources