{
  "markdown": "# Writing Skills for Claude Code\n\nA Claude Code plugin with specialized skills for fiction and narrative nonfiction authors.\n\n## Session Continuity\n\nBook projects span weeks or months, and Claude has no memory between sessions. All three skills now treat the project's blueprint or Story Bible as persistent state: read it at session start, update it when foundations change, and write a short note to `sessions/YYYY-MM-DD_topic-slug.md` at session end. The prose-mechanics skill uses a per-project audit tracker the same way.\n\n## Skills Included\n\n### Fiction Workshop\nCollaborative fiction writing and editing with 5 editorial personas:\n- **Developmental Editor** - Plot, pacing, structure, stakes\n- **Line Editor** - Prose rhythm, word choice, \"show don't tell\"\n- **Character Consultant** - Voice consistency, motivation, arc\n- **Continuity Tracker** - Timeline, world facts, internal consistency\n- **Brainstorm Partner** - \"What if\" exploration, problem-solving\n\nIncludes genre-specific guides for:\n- Spy thrillers (tradecraft, tension, moral complexity)\n- Hard sci-fi (technical accuracy, worldbuilding, geopolitics)\n\n### Character Archetypes\nFiction-only companion to Fiction Workshop's character work. Two taxonomies plus four analysis modes for using archetypes diagnostically:\n- **Narrative-Role Archetypes** - Vogler/Campbell's 8 character functions (Hero, Mentor, Threshold Guardian, Herald, Shapeshifter, Shadow, Trickster, Ally)\n- **Personality Archetypes** - Jungian 12 (Mark & Pearson model): Innocent, Everyman, Hero, Caregiver, Explorer, Rebel, Lover, Creator, Jester, Sage, Magician, Ruler\n- **Analyzer** - Diagnose or recommend an archetype pairing\n- **Audit** - Flag stock/cliché use of an established archetype\n- **Conformance** - Check for unexplained archetype drift across chapters\n- **Ensemble** - Check cast-level archetype balance and gaps\n\nArchetype is a starting scaffold, not a finished character — pair with Fiction Workshop's Want/Need/Wound/Lie framework to individualize.\n\n### Story Structure\nPercentage-anchored macro plot structure, drawn from K.M. Weiland's landmark-beat model and James Scott Bell's signpost catalog:\n- **Landmark Beats** - 11-beat skeleton (Hook, Inciting Event, First Plot Point, Pinch Points, Midpoint, Third Plot Point, Climax, Resolution) with percentages and diagnostics\n- **Signposts** - Bell's 14 named checkpoints (Disturbance, Mirror Moment, Doorways of No Return, Q Factor, and more), overlaid on the landmarks or placed relatively between them\n- **Map** - Place beats and signposts for a story not yet drafted\n- **Audit** - Locate where beats actually land in an existing manuscript and flag missing, mistimed, or thin ones\n\nWeiland's beats answer *where* a turning point falls; Bell's signposts answer *what psychological work* it needs to do.\n\n### Narrative Nonfiction\nFor self-help, prescriptive nonfiction, and structural-argument books with storytelling elements:\n- **Transformation Arc** - Reader journey design\n- **Metaphor Consistency** - Extended metaphor management\n- **Exercise Design** - Practical application sections\n- **Reveal Engineering** - Twist/reframe setup and payoff across four patterns:\n  - *Permission-reframe* (training-wheels frame released as validation)\n  - *Empirical* (counterintuitive data overturns received wisdom)\n  - *Structural* (visible surface reframed by hidden mechanism)\n  - *Conceptual* (new category installed; old framework dissolved)\n- **Voice Editing** - Tone and persona consistency\n\n### Prose Mechanics\nSentence-level diagnostic audits for finished or near-finished drafts (fiction or nonfiction):\n- **Active/Passive Audit** - Unjustified passive constructions, hidden agency\n- **Parallel Structure Audit** - Broken grammatical parallels in lists, comparisons, series\n- **Sentence Length Variance** - Flat-rhythm detection at the paragraph level\n- **Accessibility Audit** - Readability scoring and structural accessibility (prose, not WCAG)\n\nRun audits one at a time, in order. Each pass produces a flagged-issues report for author review.\n\n### Avoid AI Writing\nAudits and rewrites text to remove AI-ism patterns that make prose sound\nmachine-generated. Supports two modes:\n- **Detect mode** — Flag AI-isms only; no rewriting. Use when auditing\n  published content, someone else's writing, or when you want to decide\n  yourself what to fix.\n- **Rewrite mode** (default) — Flag AI-isms and rewrite the text to fix them.\n\nRecommended workflow: run `avoid-ai-writing` after the line-edit pass in\neither `fiction-workshop` or `narrative-nonfiction` to catch any residual\nAI patterns before final polish.\n\nVendored from [conorbronsdon/avoid-ai-writing](https://github.com/conorbronsdon/avoid-ai-writing)\n(MIT). See [ATTRIBUTION.md](ATTRIBUTION.md).\n\n## Installation\n\n### Direct from GitHub\n```bash\n/plugin install github:rhavekost/author-toolkit\n```\n\n### For Team Projects\nAdd to `.claude/settings.json`:\n```json\n{\n  \"enabledPlugins\": [\n    \"github:rhavekost/author-toolkit\"\n  ]\n}\n```\n\n## Usage\n\nInvoke skills directly:\n\n```bash\n/author-toolkit:fiction-workshop\n/author-toolkit:character-archetypes\n/author-toolkit:story-structure\n/author-toolkit:narrative-nonfiction\n/author-toolkit:prose-mechanics\n/author-toolkit:avoid-ai-writing\n```\n\nOnce activated, work with the editorial personas:\n\n```\n# Fiction\n\"As developmental editor, analyze Chapter 3\"\n\"As line editor, polish this dialogue\"\n\"Brainstorm mode—I need to solve [plot problem]\"\n\n# Character archetypes\n\"What archetype is this character?\"\n\"Audit this scene for archetype cliché\"\n\"Check archetype conformance for chapters 4-8\"\n\"Run an ensemble balance check on the cast\"\n\n# Story structure\n\"Map the structure for this story\"\n\"Audit this manuscript's structure\"\n\"Check this manuscript for a Faux Climax\"\n\n# Nonfiction\n\"Help me design the transformation arc for my self-help book\"\n\"Check metaphor consistency in chapters 4-8\"\n\"Engineer the reveal for chapter 16\"\n\n# Prose mechanics\n\"Run active/passive audit on chapter 3\"\n\"Run sentence-variance audit on chapter 7\"\n\"Run accessibility audit on the whole manuscript\"\n\n# Avoid AI Writing\n\"Audit this chapter for AI-isms\" (detect mode)\n\"Rewrite this chapter to remove AI patterns\" (rewrite mode)\n```\n\n## Quick Reference\n\n| Need | Invoke Skill | Then Say |\n|------|--------------|----------|\n| Fiction writing/editing | `/author-toolkit:fiction-workshop` | \"Let's build a story bible\" or \"As developmental editor, analyze Chapter 3\" |\n| Developmental pass | `/author-toolkit:fiction-workshop` | \"As developmental editor, analyze [chapter]\" |\n| Line editing | `/author-toolkit:fiction-workshop` | \"As line editor, polish [scene]\" |\n| Character work | `/author-toolkit:fiction-workshop` | \"As character consultant, is this in character?\" |\n| Continuity check | `/author-toolkit:fiction-workshop` | \"As continuity tracker, check for inconsistencies\" |\n| Get unstuck | `/author-toolkit:fiction-workshop` | \"Brainstorm mode—I need to [solve problem]\" |\n| Archetype identification | `/author-toolkit:character-archetypes` | \"What archetype is this character?\" |\n| Archetype cliché check | `/author-toolkit:character-archetypes` | \"Audit [scene] for archetype cliché\" |\n| Archetype drift check | `/author-toolkit:character-archetypes` | \"Check archetype conformance for chapters [X-Y]\" |\n| Cast balance check | `/author-toolkit:character-archetypes` | \"Run an ensemble balance check on the cast\" |\n| Plan new story structure | `/author-toolkit:story-structure` | \"Map the structure for this story\" |\n| Audit existing structure | `/author-toolkit:story-structure` | \"Audit this manuscript's structure\" |\n| Nonfiction writing | `/author-toolkit:narrative-nonfiction` | \"Let's build a blueprint for [book]\" |\n| Reader journey | `/author-toolkit:narrative-nonfiction` | \"Design the transformation arc\" |\n| Metaphor consistency | `/author-toolkit:narrative-nonfiction` | \"Check metaphor consistency in chapters 4-8\" |\n| Exercise design | `/author-toolkit:narrative-nonfiction` | \"Design exercises for [concept]\" |\n| Active/passive pass | `/author-toolkit:prose-mechanics` | \"Run active/passive audit on [chapter]\" |\n| Parallel structure | `/author-toolkit:prose-mechanics` | \"Run parallel-structure audit on [chapter]\" |\n| Sentence variance | `/author-toolkit:prose-mechanics` | \"Run sentence-variance audit on [chapter]\" |\n| Readability audit | `/author-toolkit:prose-mechanics` | \"Run accessibility audit on [chapter]\" |\n| Audit chapter for AI-isms | `/author-toolkit:avoid-ai-writing` | \"Detect mode — flag AI patterns in this chapter\" |\n| Clean AI-isms from chapter | `/author-toolkit:avoid-ai-writing` | \"Rewrite mode — remove AI patterns from this chapter\" |\n\n## Attribution\n\nThis plugin vendors the `avoid-ai-writing` skill by\n[Conor Bronsdon (@ConorBronsdon)](https://github.com/conorbronsdon),\nsourced from\n[conorbronsdon/avoid-ai-writing](https://github.com/conorbronsdon/avoid-ai-writing)\nand licensed under the MIT License. The upstream `LICENSE` and `README` are\npreserved unmodified inside\n[`skills/avoid-ai-writing/`](skills/avoid-ai-writing/). Full attribution and\nvendored commit SHA are recorded in [ATTRIBUTION.md](ATTRIBUTION.md).\n\n## License\n\nMIT. See [LICENSE](LICENSE). Vendored third-party skills retain their\noriginal licenses inside their skill directories — see\n[ATTRIBUTION.md](ATTRIBUTION.md) for details.\n",
  "bytes": 9332,
  "sha": "6566ab79b49e8f3facf10240d6ac509fa5ec1279f85527f1e1c33c01caf5baaa",
  "repo_slug": "rhavekost/author-toolkit",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_rhavekost_author_toolkit_fiction_worksho_38de8749/readme"
}