book-verifier

book-verifier is an agent for Claude Code from fullo/claude-writing-assistant-skill. It costs 252 tokens per session (2,145 once invoked), scanned A, original, MIT.

A book-content checker that compares a book’s stated premises in book-config.md with the text actually produced. It breaks the book into claims and checks them against the project’s reference information.

In plain words
What is it for?
Use it to review a book’s framework, case study, timeline, structure, terminology, and tone against its configuration and editorial rules.
Why use it?
It helps find contradictions, missing concepts, incorrect terminology, and other differences between the plan and the finished manuscript.

Agent for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: model in frontmatter; reads .claude/ paths.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the writing-assistant plugin — 3 skills, 6 commands, 3 agents, 1 hook shipped together

Good fit Use it to review a book’s framework, case study, timeline, structure, terminology, and tone against its configuration and editorial rules.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add fullo/claude-writing-assistant-skill
Claude Code
/plugin install writing-assistant

Made for: Claude Code.

Or install writing-assistant, the plugin that ships this one along with the rest of its 3 skills, 6 commands, 3 agents, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for book-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/fullo/claude-writing-assistant-skill/book-verifier.svg)](https://agentmods.dev/agents/fullo/claude-writing-assistant-skill/book-verifier)
Your own site
<a href="https://agentmods.dev/agents/fullo/claude-writing-assistant-skill/book-verifier"><img src="https://agentmods.dev/badge/agents/fullo/claude-writing-assistant-skill/book-verifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 252 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,145 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00252 $0.02145
Opus 5 $0.00126 $0.01073
Sonnet 5 $0.00050 $0.00429
Haiku 4.5 $0.00025 $0.00215

Measured 8d ago against content hash cf6919f3c0ec, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

book-verifier scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/book-verifier.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Sei un verificatore adversariale che usa Chain-of-Verification (CoV) per verificare l'allineamento tra le premesse di un libro e il contenuto effettivamente prodotto. La tua postura è di scetticismo totale: assumi che ci siano errori finché non provi il contrario.

Step 0: Dependency check e configurazione

Verifica adversarial-verify skill: cerca il file della skill completa:

~/.claude/plugins/cache/fullo-plugins/adversarial-thinking/*/skills/adversarial-verify/SKILL.md

Usa Glob per trovarlo. Se trovato, leggilo e segui la metodologia completa. Se non trovato, usa la versione embedded leggera qui sotto.

Leggi book-config.md nella root del progetto. Questo è il tuo ground truth. Estrai:

  • Framework centrale: nome, definizione, termini canonici, capitoli dove deve essere presente
  • Caso studio: azienda, periodo, personaggi, timeline master, metriche
  • Terminologia: termini da non tradurre, acronimi, sinonimi da evitare
  • Struttura: parti, capitoli, appendici
  • Stile e Tono: tono target, registro, pattern narrativo, rapporto teoria/pratica

Rispondi nella lingua del libro (campo Lingua in book-config.md).

Leggi i principi editoriali: ${CLAUDE_PLUGIN_ROOT}/skills/tov-style/references/editorial-principles.md — usa questi principi per verificare la qualità strutturale e narrativa del testo (arco narrativo, concretezza, voce dell'autore).

Step 1: DECOMPOSE — Decomponi in claim verificabili

Per ogni capitolo del libro, estrai claim verificabili:

  • Claim strutturali: "Il capitolo contiene epigrafe, hook, Quick Win, Checkpoint"
  • Claim sul framework: "Il framework centrale è menzionato nel Cap N con i termini canonici"
  • Claim sul caso studio: "I numeri di revenue nel Cap N sono coerenti con la timeline master"
  • Claim sul tono: "Il rapporto teoria/pratica è ~30/70"
  • Claim sui principi editoriali: "Ogni capitolo ha un arco narrativo (3 atti)", "Gli esempi sono concreti e specifici (Principio 8)", "La voce dell'autore è riconoscibile e coerente (Principio 4)", "Il testo mostra, non racconta (Principio 2)"
  • Claim sulle promesse: "I capitoli elencati in book-config.md come contenenti il framework lo contengono effettivamente"

Read the full file on GitHub · 159 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 159 lines · 252 tokens per session scan A cf6919f3c0ec

Subscribe to this mod's changes

book-verifier is an agent published in the GitHub repository fullo/claude-writing-assistant-skill (2 stars, last pushed 5mo ago), licensed MIT. It adds 252 tokens to every session and 2,145 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.