From haowjy-creative-writing-skills
Guides story structuring at saga, arc, chapter, and scene levels with focus on causation, escalating stakes, pacing, and setup-payoff. Use for planning arcs, chapter outlines, or evaluating narrative structure.
npx claudepluginhub haowjy/creative-writing-skills --plugin creative-writing-skillsThis skill uses the workspace's default tool permissions.
Structure story at multiple levels so that each scene, chapter, and arc serves the larger narrative. This skill covers structural thinking — how parts relate to wholes, how pacing works across scales, how setup creates payoff. It's not prescriptive about methodology (three-act, hero's journey, kishotenketsu, etc.) because structure should serve the story, not the other way around.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Generates original PNG/PDF visual art via design philosophy manifestos for posters, graphics, and static designs on user request.
Structure story at multiple levels so that each scene, chapter, and arc serves the larger narrative. This skill covers structural thinking — how parts relate to wholes, how pacing works across scales, how setup creates payoff. It's not prescriptive about methodology (three-act, hero's journey, kishotenketsu, etc.) because structure should serve the story, not the other way around.
Stories operate at nested scales. Decisions at one level constrain and enable decisions at others.
Saga/Series — the full story spanning multiple arcs. What's the overarching question or transformation? How do arcs build on each other? Where are the major turning points that redefine what the story is about?
Arc — a self-contained narrative movement, typically spanning multiple chapters. Each arc has its own question, rising tension, and resolution — but also advances the saga-level story. An arc that resolves its own conflict but doesn't change anything at the saga level is filler.
Chapter — a unit of reading. Chapters need internal momentum (something changes by the end) and external momentum (the reader wants to continue to the next one). Not every chapter is a complete dramatic unit — some are part of a multi-chapter sequence.
Scene — the fundamental building block. A scene happens in a specific time and place, involves specific characters, and changes something. Scenes that don't change anything are candidates for cutting or combining.
"And then" is not structure. "Therefore" and "but" are. Events should cause or complicate each other, not merely follow each other. When you outline, check whether each beat is caused by the previous one or merely comes after it.
A simple test: can you reorder the scenes without the story breaking? If yes, the structure is a sequence, not a plot.
Each structural level should raise the stakes from the previous one. The first arc's conflict should matter less than the second arc's. Within an arc, early chapters establish what could be lost; later chapters threaten it.
Escalation doesn't always mean bigger explosions. Stakes can escalate emotionally (losing a friend matters more than losing a fight), morally (the right choice becomes harder), or informationally (each revelation makes the situation more complex).
Every significant payoff needs setup, and every significant setup needs payoff. The setup doesn't need to be obvious — in fact, the best setups are invisible until the payoff arrives. But they need to exist. A twist that works relies on the reader being able to look back and see it was built.
Tracking setup/payoff across arcs is one of the hardest structural problems in long-form fiction. Maintain explicit records of open setups (foreshadowing, Chekhov's guns, unanswered questions, promises to the reader) and check them regularly.
Sustained tension becomes numbness. The reader needs breathing room — not absence of conflict, but variation in intensity. A quiet conversation after a battle. A comedic beat after an emotional gut-punch. Structure these intentionally; the rhythm of tension and release is what gives a story its emotional shape.
This applies at every scale. Within a scene: not every line should be at maximum intensity. Within a chapter: alternate high and low beats. Within an arc: the climax lands harder when preceded by a quieter stretch.
An arc is defined by a central question — something the characters (and reader) need answered. The question creates tension, the arc's events explore it from different angles, and the resolution provides an answer (which may itself raise new questions for the next arc).
Hook — what pulls the reader (and characters) into this arc? The inciting event, the new complication, the revelation that changes everything. This doesn't have to be the first scene — an arc can ease in through character moments before the central tension arrives.
Rising complications — each event makes the central question harder to answer, not easier. If the protagonist keeps winning, tension drops. Complications should be logical consequences of previous choices, not random obstacles.
Midpoint shift — somewhere in the middle, something changes the nature of the conflict. The protagonist learns something that reframes the problem. An ally becomes an antagonist. The stakes change category. Without this, arcs sag in the middle.
Crisis — the moment of maximum tension where the protagonist must make a defining choice. The choice should be genuinely difficult — not between good and bad, but between competing values or between two kinds of loss.
Resolution — the aftermath of the crisis. Not just "what happened" but "what changed." The resolution of one arc's question should set up the next arc's question.
Arcs have their own rhythm independent of chapter structure. An arc might span 5 chapters or 15. The structural beats above aren't evenly distributed — rising complications take the most space, the crisis is often compressed, and the resolution can be brief.
When pacing feels off, check whether the arc's structural beats are landing where they need to. A midpoint shift in chapter 2 of a 10-chapter arc means the second half has to sustain tension without its primary tool for renewal.
Chapters are reading units, not structural units. A chapter can contain multiple scenes or one long scene. It can advance the arc's plot or develop character while the plot pauses. What makes it work as a chapter is internal completeness and external pull.
Internal completeness — something should change within the chapter. The reader should feel they received something — a revelation, an emotional shift, a new complication, a moment of connection. Chapters that are purely transitional ("getting from A to B") feel like filler.
External pull — the chapter should end with forward momentum. An unanswered question, a new threat, an emotional shift that demands exploration. This doesn't mean every chapter needs a cliffhanger — subtler forms of pull work too. The reader should want to know what happens next.
A scene is the smallest structural unit: a continuous sequence in a specific time and place. Every scene should earn its presence.
Questions to ask about a scene:
Scenes that only exist to convey information can usually be cut — find a scene that's already doing something and deliver the information there.
Some writers plan extensively before drafting. Some discover the story by writing it. Most do some of both. This skill doesn't advocate for either approach — it provides structural thinking that applies regardless.
For planners: outlines are hypotheses, not contracts. If the draft reveals that the outline's structure doesn't work, change the outline. Serving the outline instead of the story produces technically correct but emotionally dead fiction.
For discovery writers: structural awareness during drafting helps you recognize when you've wandered into a dead end. You don't need a detailed outline, but knowing what your current arc's central question is helps you evaluate whether a scene is exploring that question or avoiding it.
When structure gets complex — parallel timelines, multiple POV threads, arc overlap — visual representations help. Mermaid diagrams can map:
Use diagrams when the structure is too complex to hold in your head, not as decoration.
Saggy middle. The arc's complications aren't actually complicating — they're repetitive variations of the same obstacle. Fix by adding a midpoint shift that changes the nature of the conflict.
Rushed ending. The resolution doesn't have space to land. Fix by either extending the resolution or simplifying what needs resolving.
Filler arcs. An arc that resolves internally but doesn't advance the saga. Every arc should change the story's permanent state — relationships, power dynamics, character understanding, stakes.
Disconnected scenes. Scenes that are individually interesting but don't connect causally. "This happened, then this happened" instead of "this happened, which caused this."
Stakes plateau. The story's stakes stop escalating. Later arcs should matter more than earlier ones — not necessarily bigger, but deeper. If arc 3 feels less consequential than arc 1, the structure is deflating.