skill CX Codex

creative-writing-muse — Claude Code skill: writing with one agent

Q: How do I install creative-writing-muse in Claude?

Run `npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a claude-code`. The skill lands at `~/.claude/skills/creative-writing-muse/SKILL.md`.

Q: Why doesn't the skill trigger on its own?

Its frontmatter sets `model-invocable: false`, so the model won't invoke it automatically. Load it manually when you sit down to write.

A skill for a single agent that switches stances on its own: plans, drafts, critiques and remembers. Useful when no subagents are around.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓

npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a claude-code

Open on GitHub

What creative-writing-muse does

This skill loads when there are no subagents and one agent has to do everything. It acts as the muse: keeps the author-facing thread coherent while moving between direction, drafting, critique, revision and memory. First it works out author intent — the reader simulation you want, the emotional target, constraints, and what should stay unsaid. That intent stays visible through every stance. The author has the final say.

Before each stance the agent names the prompt it gives itself: what this pass is for, which reader effect it protects, which canon and style details matter now, what should stay rough or strange. It asks the author only when the answer would change the work. Otherwise it states its read and continues.

Stances stay separate. Explore without committing early, draft before judging, critique from the reader's experience, then revise the highest-impact issue. Memory updates only for settled facts and decisions. For pivotal passages the skill suggests two meaningfully different takes and an explanation of what each proves. One catch: the skill is marked model-invocable: false, so the model will not pick it up by itself — you load it by hand.

What it helps with

run the whole writing cycle with one agent and no subagents
keep author intent visible across stance changes
produce two different takes on a pivotal passage
separate drafting from critique instead of judging early
store only settled facts in story memory

Example prompts for the agent

/creative-writing-muse help me plan this chapter and draft it
/creative-writing-muse critique this scene from the reader's experience
/creative-writing-muse give me two takes on the ending and what each proves

How to install creative-writing-muse

1Skills CLI (Vercel)

npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a claude-code

2Manually: put the files in place

~/.claude/skills/creative-writing-muse/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/creative-writing-muse/

1Skills CLI (Vercel)

npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a codex

2Manually: put the files in place

.agents/skills/creative-writing-muse/SKILL.md

For all projects use ~/.agents/skills/

1Skills CLI (Vercel)

npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a cursor

2Manually: put the files in place

.cursor/skills/creative-writing-muse/SKILL.md

Cursor also reads .agents/skills/ and .claude/skills/

1GitHub CLI

gh skill install haowjy/creative-writing-skills creative-writing-muse

2Manually: put the files in place

.github/skills/creative-writing-muse/SKILL.md

1Skills CLI (Vercel)

npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a gemini-cli

2Platform CLI

gemini skills install https://github.com/haowjy/creative-writing-skills --path skills/creative-writing-muse

Manually: put the files in place

.devin/skills/creative-writing-muse/SKILL.md

The legacy .windsurf/skills/ path still works

Skills CLI (Vercel)

npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a opencode

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about creative-writing-muse

How do I install creative-writing-muse in Claude?

Run npx skills add haowjy/creative-writing-skills --skill creative-writing-muse -a claude-code. The skill lands at ~/.claude/skills/creative-writing-muse/SKILL.md.

What does creative-writing-muse do?

It gives one agent a mode-shift routine: direction, drafting, critique, research and memory. Author intent carries across every pass.

Why doesn't the skill trigger on its own?

Its frontmatter sets model-invocable: false, so the model won't invoke it automatically. Load it manually when you sit down to write.

Source text

SKILL.md · Apache-2.0

---
name: creative-writing-muse
type: mode-shift
description: >
  Load when no subagents are available and one agent must plan, draft,
  critique, research, and capture memory by switching stances.
model-invocable: false
---

# Creative Writing Muse

Use this when there are no subagents. Act as the muse in one conversation by
loading the relevant writing skills and switching stances deliberately. Keep the
author-facing thread coherent while you move between direction, drafting,
critique, revision, and memory.

Start by understanding author intent: desired reader simulation, emotional
target, constraints, taste signals, open uncertainty, and what should remain
unsaid. Keep that intent visible as you change stance. The author has the final
say.

Before writing character prose, find the style and character state.

## Choose the Stance

Load the skills needed for the next stance:

- **Direction:** `/story-planning`
- **Drafting:** `/creative-writing-modes`, `/creative-writing-craft`, `/llm-writing`
- **Critique:** `/story-review`, `/reader-sim`, `/writing-principles`
- **Research:** `/creative-research`
- **Voice and terms:** `/creative-writing-craft`, `/character-sim`, `/shared-dao`
- **Memory:** `/story-memory`; also `/kb-management` and `/project-setup` if available

## Self-Prompt Before Each Stance

Before doing the next pass, name the prompt you are giving yourself:

- What is the author's intent for this pass?
- What reader effect should the output create or protect?
- Which constraints, style, character state, canon, and vocabulary matter now?
- What should remain ambiguous, unresolved, rough, or strange?
- What output should this pass produce?
- What would be the wrong kind of success?

Ask the author only when the answer would change the work. Otherwise state your
read and continue.

## Keep Stances Separate

Explore without committing too early. Draft before judging. Critique from the
reader's experience. Revise the highest-impact issue. Update memory only for
settled facts and decisions.

Before switching stance, synthesize what changed and whether the next move still
serves the author's intent. For pivotal passages, create two meaningfully
different takes and explain what each take proves.

Similar skills

Everything for Writing and docs →

academic-writing poemswe A skill for academic writing that strips template phrasing and pushes for names, numbers and cited evidence instead of generalities. CX Codex +1 1,000installs better-writing forjd Rewrites, drafts and reviews prose for emails, reports, docs and UI copy. Useful when text reads generic, salesy or like a chatbot wrote it. CX Codex +1 514installs writing-guidelines vercel A skill that reviews docs and prose against a written style guide. You get findings with line numbers instead of hunting through the file yourself. CX Codex +1 107,100installs article-writing affaan-m Drafts and edits articles, guides, tutorials, and newsletters in a specified voice. Use it when a long piece needs a clear structure and factual support. CX Codex +1 5,200installs avoid-ai-writing Conor Bronsdon Audits and rewrites prose to remove machine-generated writing patterns. Useful when editing drafts or documentation to eliminate characteristic AI phrasing. CX Codex +1 4,600installs humanize-writing jpeggdev Rewrites AI-generated drafts by stripping formulaic structure, puffery, and recognizable model phrasing. CX Codex +1 2,000installs

From the same set22

intent-modeling A skill that makes an AI agent work out what you meant before it acts on what you literally said. CX Codex +1 527installs story-memory A skill that keeps canon, facts and context alive across chapters. Useful once a fiction project outgrows a single session. CX Codex +1 527installs shared-dao Keeps story terms consistent across lore files and drafts, organizing them inside kb/. It eliminates drifting names, ambiguous factions, and conflicting magic systems. CX Codex +1 520installs grill-with-docs Challenges project plans through sequential questioning, verifies terminology against project docs, and records decisions on the fly. CX Codex +1 514installs creative-research This skill gathers grounded material for a story question: history, culture, daily detail, how a trade actually works. You get back a sourced report. CX Codex +1 487installs information-hierarchy A skill for large human-facing output: reports, docs, multi-section explanations. The answer comes first, depth sits behind it. CX Codex +1 399installs structured-artifact A skill for building structured HTML artifacts: the agent decides how to present the content first, then picks the tools. CX Codex +1 387installs zoom-out Explores an unfamiliar area from several angles at once, then folds the findings into a single orientation map. CX Codex +1 386installs