agentleFS
ek0212ek0212/agent-ready-repo-kit/skills/creative-writing/SKILL.md

creative-writing

Skill0 starsChanged 4 months agoMITView on GitHub

Draft, structure, or edit narrative nonfiction with strong questions, scenes, characters, reveals, and endings. Use when user asks for a blog post, article, profile, essay, narrative report, story-driven rewrite, title options, or stronger nonfiction storytelling.

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 8 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write narrative nonfiction from verified material.

    line 7
  2. Very specific:

    Use for articles, profiles, essays, blogs, and narrative reports.

    line 11
  3. Distinctive:

    Write one central question.

    line 26
  4. Distinctive:

    Avoid more than two argument-heavy sections without concrete relief.

    line 35
  5. Distinctive:

    Use suspense when reader knows answer is coming.

    line 43
  6. Distinctive:

    Delete sections that do not serve central question.

    line 56
  7. Distinctive:

    Verify each factual statement against source material.

    line 58
  8. Distinctive:

    Every scene and quote sourced.

    line 72
  9. Distinctive:

    Use direct simple sentences.

    line 81

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Write narrative nonfiction from verified material. Build reader movement, not ornamental prose.

Scope

Use for articles, profiles, essays, blogs, and narrative reports. Use no-ai-slop for AI-pattern detection or human-voice polish when narrative structure is not main job. Do not use for fiction, product microcopy, straight technical docs, or research synthesis unless user asks for narrative treatment.

Inputs

  • Topic or draft.
  • Audience, venue, length, and voice.
  • Notes, interviews, quotes, data, and sources.
  • Required claims, exclusions, and call to action.

User owns facts and voice. Never invent quotes, scenes, motives, credentials, or data. Mark missing evidence with clear placeholders.

Workflow

Step 1: Pin Question And Ending

Write one central question. Name what stays unresolved. Choose ending image, fact, decision, or insight before drafting. If ending only repeats thesis, find stronger destination.

Step 2: Map Sections

Give each section one job. Alternate:

  • Argument: claim, evidence, analysis.
  • Story: scene, person, vivid fact, useful detour.

Avoid more than two argument-heavy sections without concrete relief. Short piece may use simple hook, claim, evidence, close.

Step 3: Build People Through Evidence

Show character through action, setting, dialogue, constraint, and choice. Pick two or three details that affect story. Remove decorative description.

Step 4: Control Reveals

Use suspense when reader knows answer is coming. Use surprise when new fact changes frame. Withhold only when delayed fact improves meaning without misleading reader.

Step 5: Translate Complexity

Explain jargon once. Break data into small units. Use analogy only when mapping is accurate and clarifies mechanism. For long pieces, test one analogous world; drop it if it adds detour without insight.

Step 6: Match Voice And Emotion

Match source voice, audience, and stakes. Use simple language for complex ideas. Let quotes and observed detail carry emotion. Avoid announcing humor, overstating sadness, or manufacturing certainty.

Step 7: Tighten And Title

  1. Cut repeated claims and throat-clearing.
  2. Delete sections that do not serve central question.
  3. Fix sentence rhythm by reading aloud.
  4. Verify each factual statement against source material.
  5. Confirm ending changes reader position.
  6. Draft at least three specific title options.

Fallbacks

  • Thin evidence: produce outline or questions, not invented draft.
  • Short format: skip characters and analogous world unless essential.
  • Institutional stakes: favor clarity, attribution, and restraint over suspense.

Verify

  • Central question clear.
  • Structure moves toward ending.
  • Every scene and quote sourced.
  • Claims traceable.
  • Jargon decoded.
  • Voice fits audience.
  • Title names real tension or value.
  • Draft reads naturally aloud.

Writing

Use direct simple sentences. Skip rhetorical flourishes. Avoid X-not-Y contrasts, self-answering rhetorical questions, em dashes, alliterative three-part phrases, vague inspirational pivots, unsourced claims, and unverified quotes.

Output

Return requested draft or edit. Include title options. State placeholders, assumptions, and source gaps.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.