Package Exports

@demig0d2/skills
@demig0d2/skills/bin/cli.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@demig0d2/skills) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

@demig0d2/skills

Claude skill suite by Vivid (Dheelep N)

A collection of production-grade Claude skills for book writing, humanizing AI text, upgrading skill files, and the full book production pipeline — installable in one command.

Install

# Install a single skill
npx @demig0d2/skills install book-writer

# Install everything at once
npx @demig0d2/skills install all

Skills are copied to ~/.claude/skills/ — Claude picks them up automatically on next session start.

Commands

npx @demig0d2/skills install <skill>   # Install one skill
npx @demig0d2/skills install all       # Install all skills
npx @demig0d2/skills remove <skill>    # Remove a skill
npx @demig0d2/skills remove all        # Remove all demigod skills
npx @demig0d2/skills list              # List available skills + install status
npx @demig0d2/skills status            # Show which skills are installed
npx @demig0d2/skills info <skill>      # Show details about a skill
npx @demig0d2/skills version           # Show package version

Skills

`book-writer`

Full end-to-end book authoring pipeline. Takes a concept document from seed idea to KDP-ready DOCX. Includes concept expansion, structure options, 26-question intake, research bank, table of contents, chapter writing (guided or autonomous), 7-dimension chapter audit, humanizer pass, continuity tracking, and KDP-formatted output.

npx @demig0d2/skills install book-writer

`humanizer`

Strips all 25 documented AI writing patterns AND actively rewrites toward Vivid's style DNA simultaneously. Includes a DNA protection layer that identifies and exempts authentic signature constructions before the pattern elimination pass runs. Produces: draft → self-audit → final version → changes summary.

npx @demig0d2/skills install humanizer

`overhaul`

Upgrades any skill file from its current version to a dramatically better one. Two modes:

Automated — diagnoses across 6 dimensions and rebuilds completely, no user input needed
Personalized — user-guided, decides scope/depth at each stage

npx @demig0d2/skills install overhaul

`concept-expander`

Turns a rough book seed (a paragraph, title, or bullet points) into a complete concept document the book-writer can consume. Outputs logline, core question, reader profile, emotional arc, thematic clusters, comp titles, and chapter territories.

npx @demig0d2/skills install concept-expander

`chapter-auditor`

Scores chapters across 7 dimensions before the humanizer pass: voice authenticity, pain-to-transformation arc, sentence rhythm, metaphor quality, structural integrity, AI pattern contamination, and reader resonance. Produces a PASS / CONDITIONAL PASS / REVISE verdict.

npx @demig0d2/skills install chapter-auditor

`continuity-tracker`

Maintains a running continuity log across all chapters. Tracks established facts, metaphors, insights, tone decisions, and open threads. Fires a continuity brief before each chapter and updates the log after — detects and surfaces conflicts before they compound.

npx @demig0d2/skills install continuity-tracker

`research-aggregator`

Builds a Research Bank organized by chapter — philosophical anchors, quotes, empirical findings, and conceptual oppositions. Tuned specifically for Vivid's genres: philosophy/essay, self-help/motivational, non-fiction/technical.

npx @demig0d2/skills install research-aggregator

Requirements

Node.js 16+
Claude.ai (any paid plan) or Claude Code

Skills install to ~/.claude/skills/ and are picked up automatically by Claude on session start.

Author

Dheelep N (Vivid) — developer, author, security researcher
8 published books · 6 research papers · 9+ production tools

License

MIT