Skip to content

nowzero1702/authorkit

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
.claude-plugin
.claude-plugin
 
 
docs
docs
 
 
skills-ko
skills-ko
 
 
skills
skills
 
 
templates
templates
 
 
.gitignore
.gitignore
 
 
LICENSE
LICENSE
 
 
README.md
README.md
 
 

Repository files navigation

authorkit

한국어 | 中文 | 日本語 | Русский | Français | Deutsch | Español | Português | العربية | हिन्दी | Türkçe | Italiano

Book authoring workflow skills for Claude Code.

Whether you're writing a university textbook, a practical how-to guide, or translating and adapting a reference book — authorkit systematizes the repetitive tasks of book writing: reference analysis, manuscript proofreading, diagram creation, style/terminology verification, and structure reorganization.

Who is this for?

  • Professors/Authors writing academic textbooks with reference books
  • Technical writers creating practical guides (e.g., "Data Analysis with Python")
  • Translators/Adapters localizing foreign textbooks for domestic audiences
  • Self-publishers iterating on their manuscripts

Installation

/plugin marketplace add nowzero1702/authorkit
/plugin install authorkit@nowzero1702-authorkit

Default plugin is Korean. For English version:

/plugin install authorkit-en@nowzero1702-authorkit

Update

/plugin marketplace update nowzero1702-authorkit
/reload-plugins

Quick Start

/authorkit-init

A questionnaire md file is generated. Fill in your answers in your IDE, then run the command again.

Skills & Examples

/authorkit-init — Project Initialization

Sets up your authoring project by generating a questionnaire md file.

You say:

"I'm writing an Operating Systems textbook for 2nd-year undergrads. I have Silberschatz's OS Concepts as a reference."

authorkit does:

  1. Creates authorkit/setup-questionnaire.md: 
    ## 1. Work Type
- [x] Both (reference + existing manuscript improvement)

## 3. Target Audience
> Answer: 2nd-year undergraduate students, first time learning OS concepts

## 5. Reference Materials
> Answer: C:\Users\...\references
  2. After you fill it in and re-run, generates:
    • authorkit/constitution.md — Your writing style rules
    • authorkit/glossary.md — Terminology glossary
    • authorkit/structure.md — Table of contents
    • authorkit/drafts/ch01/, ch02/, ... — Chapter folders

/authorkit-analyze — Reference/Manuscript Analysis

Extracts structure, figures, key concepts, and issues from any supported file.

You say:

"Analyze the Silberschatz reference book"

authorkit outputs (authorkit/references/ref-001/analysis.md):

## Structure
| Location | Title | Level | Volume |
|----------|-------|:-----:|:------:|
| p.1 | Ch1. Introduction | H1 | 42 pages |
| p.43 | Ch2. Operating-System Structures | H1 | 38 pages |
| p.81 | Ch3. Processes | H1 | 45 pages |
...

## Figures
| ID | Page | Caption | Extracted |
|----|------|---------|:---------:|
| FIG-001 | p.6 | Figure 1.1 Abstract view of system components | ✅ |
| FIG-002 | p.23 | Figure 1.12 Memory layout for a multiprogramming system | ✅ |
...

## Key Terminology
| Term | Abbreviation | First Appearance |
|------|:------------:|-----------------|
| Process Control Block | PCB | Ch3 |
| Context Switch | - | Ch3 |
...

## Issues Found
| Type | Location | Detail |
|------|----------|--------|
| Empty section | Ch7.4 | Heading only, no content |

Supported formats: pdf, docx, txt, xlsx, hwpx, md

/authorkit-compare — Reference ↔ Manuscript Comparison

Compares your manuscript against references to decide what to adopt, skip, or create originally.

You say:

"Compare my Ch3 with Silberschatz Ch3"

authorkit outputs (authorkit/drafts/ch03/compare.md):

## Structure Comparison
| Reference (Silberschatz) | My Manuscript | Mapping |
|-------------------------|---------------|:-------:|
| 3.1 Process Concept | 3-1 프로세스 개념 | Match |
| 3.2 Process Scheduling | 3-2 프로세스 스케줄링 | Match |
| 3.3 Operations on Processes | (missing) | Gap |
| 3.4 Interprocess Communication | 3-4 IPC | Match |

## Content Classification
### Adopt (from reference)
| Reference | Content | Target | Method |
|-----------|---------|--------|--------|
| §3.1 | Process state diagram | 3-1 | Simplify for beginners |
| §3.2 | Scheduling queue figures | 3-2 | Recreate as text block diagram |

### Skip
| Reference | Content | Reason |
|-----------|---------|--------|
| §3.3.2 | Chrome multi-process case study | Too specific, audience won't relate |

### Original Content
| Section | Content | Reason |
|---------|---------|--------|
| 3-1 intro | "Why do we need processes?" motivation | Not in reference, needed for beginners |

## Coherence Issues
| Type | Location | Detail | Severity |
|------|----------|--------|:--------:|
| Forward ref | 3-2 "as we learned in scheduling" | Scheduling not yet taught | High |

/authorkit-draft — Section-Level Drafting/Proofreading

Extracts a section as old.md, rewrites/proofreads it as new.md.

You say:

"Proofread section 3-1"

authorkit creates two files you can compare side-by-side in your IDE:

drafts/ch03/3-1-old.md (extracted from manuscript):

## 3-1 프로세스 개념

프로세스는 실행중인 프로그램입니다. 프로세스에는 여러 상태가 있는데
New, Ready, Running, Waiting, Terminated 가 있습니다...

drafts/ch03/3-1-new.md (proofread):

## 3-1 프로세스 개념

프로세스는 실행 중인 프로그램입니다. 프로세스에는 다섯 가지 상태가 있으며,
생성(New), 준비(Ready), 실행(Running), 대기(Waiting), 종료(Terminated)로
나뉩니다...

Change summary:

Type Change Count
Terminology "실행중인" → "실행 중인" 3
Terminology format "New, Ready" → "생성(New), 준비(Ready)" 5
Spacing "프로그램입니다.프로세스" → "프로그램입니다. 프로세스" 2

4 proofreading levels:

"Just fix formatting in 3-1"        → Level 1: formatting only
"Enhance 3-1 with diagrams"         → Level 2: add diagrams, explanations
"Restructure 3-1"                   → Level 3: add/remove/reorder subsections
"Rewrite 3-1 from scratch"          → Level 4: full rewrite

/authorkit-juice — Juice into Markdown

Converts reference or manuscript files to clean markdown with embedded images, dramatically reducing token consumption. Enhanced with AI-powered extraction features inspired by opendataloader-pdf.

You say:

"Juice the reference PDF to markdown"

authorkit does:

  1. Page triage — classifies each page (clean text → local, scanned → OCR, complex tables → AI)
  2. Text extraction — with reading order correction (XY-Cut algorithm) for multi-column layouts
  3. Table extraction — dual strategy: border-based (visible lines) + cluster-based (borderless tables)
  4. Formula extraction — math fonts and symbols converted to LaTeX ($E=mc^2$)
  5. Image extraction — with optional AI-generated alt text descriptions
  6. Content safety — hidden text detection (prompt injection defense), optional PII sanitization

Output (authorkit/converted/ref-001/):

full.md            ← Complete converted file
full.json          ← Structured JSON with bounding boxes (optional)
ch01.md ~ ch12.md  ← Per-chapter files
images/            ← Extracted images
conversion-log.md  ← Metadata: tables, formulas, OCR pages, processing stats

Options:

"Juice with OCR — it's scanned"              → ocr: force
"Extract borderless tables too"               → table_method: cluster
"Output structured JSON with bounding boxes"  → format: json
"Preserve math formulas as LaTeX"             → formula: on
"Generate AI descriptions for figures"        → image_desc: on
"Sanitize personal information"               → sanitize: on
Option Values Default
format markdown, json, html, text markdown
table_method auto, border, cluster auto
reading_order auto, xycut, none auto
ocr auto, force, off auto
formula on, off on
image_desc on, off off
sanitize on, off off
hidden_text warn, strip, off warn

/authorkit-diagram — Text Block Diagram Creation

Generates text block diagrams in markdown code blocks.

You say:

"Create a diagram for the process state transition"

authorkit outputs:

[Figure] Process State Transition Diagram (Silberschatz Fig 3.2)

                    admitted              interrupt
  ┌─────┐  ──────────────▶  ┌───────┐  ──────────▶  ┌─────────┐
  │ New │                    │ Ready │               │ Running │
  └─────┘                    └───┬───┘  ◀──────────  └────┬────┘
                                 │      scheduler         │
                                 │      dispatch          │
                                 │                        │
                            I/O or                   exit │
                         event completion                 │
                                 │                        ▼
                            ┌────┴────┐            ┌────────────┐
                            │ Waiting │            │ Terminated │
                            └─────────┘            └────────────┘

4 modes:

"Convert Fig 3.2 to text diagram"       → Reference-based
"Find diagrammable parts in 3-1"        → Scan mode
"Create diagram for the PCB structure"   → Content-based
"Fill in the [Figure] placeholder"       → Auto-fill mode

/authorkit-review — Style/Terminology/Structure Verification

Verifies your manuscript against constitution and glossary rules.

You say:

"Review chapter 3"

authorkit outputs (authorkit/reviews/review-2025-03-25.md):

# Review Report — Chapter 3

## Summary
| Area | Violations | High | Medium | Low |
|------|:----------:|:----:|:------:|:---:|
| Style | 8 | 0 | 3 | 5 |
| Terminology | 5 | 2 | 1 | 2 |
| Cross-references | 2 | 1 | 1 | 0 |
| Coherence | 1 | 0 | 0 | 1 |
| **Total** | **16** | **3** | **5** | **8** |

## High Severity (fix immediately)
| Location | Found | Expected (glossary) | Type |
|----------|-------|---------------------|------|
| 3-1 line 12 | "프로세스Process" | "프로세스(Process)" | Terminology format |
| 3-2 line 45 | "컨텍스트 스위치" | "문맥 교환(Context Switch)" | Wrong translation |

## Cross-Reference Issues
| Location | Reference | Actual | Status |
|----------|-----------|--------|:------:|
| 3-4 "Ch5에서 설명" | Ch5 Memory | Ch5 exists | ✓ |
| 3-2 "앞서 배운 스케줄링" | Prior section | Not yet taught | ✗ |

## Auto-fixable
The following can be fixed with `/authorkit-draft 3-1 formatting`:
- 5 spacing issues
- 3 terminology format issues

/authorkit-restructure — Structure Reorganization

Analyzes section ordering and suggests restructuring based on prerequisite dependencies.

You say:

"Analyze chapter 3 structure"

authorkit outputs:

## Dependency Graph

  3-1 Process Concept
       │
       ├──▶ 3-2 Process Scheduling
       │         │
       │         ▼
       │    3-4 CPU Scheduling
       │
       ▼
  3-3 IPC ──▶ 3-5 Synchronization

## Coherence Violations
| Current Order | Issue |
|---------------|-------|
| 3-4 before 3-2 | 3-4 references scheduling queues from 3-2 |

## Suggested Reordering
| Current | Suggested | Reason |
|---------|-----------|--------|
| 3-1 | 3-1 (keep) | Foundation |
| 3-3 IPC | → 3-2 | Independent of scheduling |
| 3-2 Scheduling | → 3-3 | Prerequisites for 3-4 |
| 3-4 CPU Scheduling | → 3-4 (keep) | Depends on 3-3 |
| 3-5 Sync | → 3-5 (keep) | Depends on IPC |

Impact: 4 cross-references need updating.

Run `/authorkit-restructure apply` to execute.

Supported File Formats

Format Read Notes
.pdf ✅ Structure, figures, text extraction
.docx ✅ Headings, styles, images
.txt ✅ Plain text
.xlsx ✅ Sheet-based content
.hwpx ✅ Korean word processor format
.md ✅ Markdown headings

Workflow

/authorkit-init          Set up project (questionnaire → constitution → glossary)
       ↓
/authorkit-analyze       Analyze references and manuscript
       ↓
/authorkit-compare       Compare reference ↔ manuscript
       ↓
/authorkit-juice       Juice files into clean markdown (token savings)
       ↓
/authorkit-draft         Write/proofread sections (old → new)
       ↓
/authorkit-diagram       Create text block diagrams
       ↓
/authorkit-review        Verify style, terminology, cross-references
       ↓
/authorkit-restructure   Reorganize chapter/section order

Language Versions

Plugin Language Install
authorkit 한국어 (default) /plugin install authorkit@nowzero1702-authorkit
authorkit-en English /plugin install authorkit-en@nowzero1702-authorkit

License

MIT

About

마감에 쫒기는 작가님들을 위한 spec-kit 기반 author-kit

Resources

Readme

License

MIT license
Activity

Stars

1 star

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors