Image 1 of 1
ChatGPT/LLM Prompt: Chapter/Scene Brief (Writing and Editing Books)
This is a single, copy/paste planning prompt for ChatGPT (or any LLM) that creates a “Chapter/Scene Brief” for one unit of your book (a chapter, section, or scene). It turns your outline (and optional Book Blueprint/canon) into a practical drafting blueprint so you can write faster, stay consistent, and avoid plot/logic drift.
It’s designed for:
fiction: planning a chapter or scene with clear beats, escalation, emotional shift, and a strong ending hook
nonfiction: planning a chapter/section with a clear promise, logical flow, examples, objections, and an optional reader exercise
Why it’s useful
You plan one unit at a time, so drafting stays focused and manageable.
It enforces continuity: you can paste your canon facts, terminology, and timeline anchors so the unit doesn’t contradict earlier material.
It’s highly customizable but simple to run: fill in what you know, leave the rest blank.
Dialogue is controlled by a percentage input (0–100). Default 0 means it plans for no dialogue.
It can ask a user-controlled number of specific clarifying questions first (0–10), then still outputs the brief using reasonable assumptions (no waiting).
What the output looks like (plain text, copy/paste friendly)
When you run the prompt, you get a structured brief like this:
Questions (optional; Q1..Qn)
BRIEF START
Assumptions Made (optional)
Unit Identification (type, identifier, placement, purpose)
Context Snapshot (what must be true before/after; canon constraints)
Voice and Style Targets (tone, pacing, style traits, what to emphasize/avoid)
Unit Structure Plan
Fiction: POV/setting, beats in order, escalation, dialogue plan (optional), ending hook
Nonfiction: promise/outcome, numbered flow of points, evidence needs, reader action, transition
Required Details (must appear in the draft)
Continuity and Consistency Checks (pre-draft checklist)
Risk List (what could go wrong + how to prevent it)
BRIEF END
If the output is long and hits limits, it ends with: CONTINUE
This is a single, copy/paste planning prompt for ChatGPT (or any LLM) that creates a “Chapter/Scene Brief” for one unit of your book (a chapter, section, or scene). It turns your outline (and optional Book Blueprint/canon) into a practical drafting blueprint so you can write faster, stay consistent, and avoid plot/logic drift.
It’s designed for:
fiction: planning a chapter or scene with clear beats, escalation, emotional shift, and a strong ending hook
nonfiction: planning a chapter/section with a clear promise, logical flow, examples, objections, and an optional reader exercise
Why it’s useful
You plan one unit at a time, so drafting stays focused and manageable.
It enforces continuity: you can paste your canon facts, terminology, and timeline anchors so the unit doesn’t contradict earlier material.
It’s highly customizable but simple to run: fill in what you know, leave the rest blank.
Dialogue is controlled by a percentage input (0–100). Default 0 means it plans for no dialogue.
It can ask a user-controlled number of specific clarifying questions first (0–10), then still outputs the brief using reasonable assumptions (no waiting).
What the output looks like (plain text, copy/paste friendly)
When you run the prompt, you get a structured brief like this:
Questions (optional; Q1..Qn)
BRIEF START
Assumptions Made (optional)
Unit Identification (type, identifier, placement, purpose)
Context Snapshot (what must be true before/after; canon constraints)
Voice and Style Targets (tone, pacing, style traits, what to emphasize/avoid)
Unit Structure Plan
Fiction: POV/setting, beats in order, escalation, dialogue plan (optional), ending hook
Nonfiction: promise/outcome, numbered flow of points, evidence needs, reader action, transition
Required Details (must appear in the draft)
Continuity and Consistency Checks (pre-draft checklist)
Risk List (what could go wrong + how to prevent it)
BRIEF END
If the output is long and hits limits, it ends with: CONTINUE
