Book Write — Testing skill for Claude Code
Draft, review, and finalize one chapter — full pipeline with a pre-draft validity check, the seven-check QA gate, Override Contracts for soft and gating findings, a revision loop with hard caps, and e.
How to install Book Write
Installs to ~/.claude/skills/jayping23-book-forge-book-write/SKILL.md
mkdir -p ~/.claude/skills/jayping23-book-forge-book-write && curl -fsSL https://raw.githubusercontent.com/JayPing23/book-forge/HEAD/commands/book-write.md -o ~/.claude/skills/jayping23-book-forge-book-write/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Book Write does
description: Draft, review, and finalize one chapter — full pipeline with a pre-draft validity check, the seven-check QA gate, Override Contracts for soft and gating findings, a revision loop with hard caps, and escalation on exhausted retries argument-hint: [--compete=N]
book-write
Drives one chapter through the full pipeline as a persistent state machine. State lives at `projects//.project-memory/chapter-state/.json` — read it f
Alternatives in Testing
- Deslop Reborn — One iteration of the IronClaw Reborn de-slop loop — take ONE Reborn crate, fan out parallel review sub-agents 12.6k ★
- Paperjury — Pre-submission AI review stress-test for research papers 1k ★
- Spec Draft PR — Turn a feature idea or GitHub issue into a one-page spec saved to Documentation/specs/ and opened as a draft P 324 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Cleaning Pass
Stage 2 cleaning pass - watermark/artifact removal via the detect -> gate -> mask -> LaMa inpaint -> verify lo
Book Export
Build an upload-ready publishing package — web-novel platform package (listing + tags + cover spec + scheduled
Book Revise
Revise an already-finalized chapter against your own note — re-enters the full QA pipeline with your feedback
Aegis Coverage Screen
Tool-boundary screen — warn user about ANY project where AEGIS coverage < 100%, list every gap. Mandatory at /
Myflow Controller
A controller for AI coding agents that takes 'done' away from the model: a Claude Code skill plus a stdlib-onl
Fix And Validate
Run the Lancet2 fix-mode validation suite (iwyu-fix + lint-check + test). Mirrors the pre-commit gate's scope
Related Agents
Outline Adherence Reviewer
One of book-forge's seven QA reviewers. Diffs a drafted chapter against its planned scene-level outline beats,
Code4rena Judge
Validates smart contract security findings against Code4rena audit competition standards. Determines correct s
Venue Suitability Scorer
Rank dining venues for company dinners and group events. Use when verified candidates need scoring by hard con