spec-auditor

spec-auditor is an agent for coding agents from queelius/claude-anvil. It costs 44 tokens per session (851 once invoked), scanned A, original, MIT.

An internal checking agent that compares a drafted book section with its chapter plan.

In plain words
What is it for?
Use it as part of Bookwright's drafting and review workflow to produce an audit report with evidence about which plan requirements the draft meets.
Why use it?
It identifies missing required content, labels, notation, cross-references, or page-length targets before the section is considered finished.

Agent

Part of the bookwright plugin — 3 skills, 11 commands, 11 agents, 1 hook shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/queelius/claude-anvil/spec-auditor
Clone the repo
git clone --depth 1 https://github.com/queelius/claude-anvil

Or install bookwright, the plugin that ships this one along with the rest of its 3 skills, 11 commands, 11 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 spec-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/queelius/claude-anvil/spec-auditor.svg)](https://agentmods.dev/agents/queelius/claude-anvil/spec-auditor)
Your own site
<a href="https://agentmods.dev/agents/queelius/claude-anvil/spec-auditor"><img src="https://agentmods.dev/badge/agents/queelius/claude-anvil/spec-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 851 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00044 $0.00851
Opus 5 $0.00022 $0.00426
Sonnet 5 $0.00009 $0.00170
Haiku 4.5 $0.00004 $0.00085

Measured 4d ago against content hash beee577758a9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

spec-auditor 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 4d 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.

bookwright/agents/spec-auditor.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You verify that a drafted section satisfies its per-chapter plan specification. You produce a structured audit report. You do not edit any file.

Inputs

You receive in the dispatch prompt the section identifier (e.g., "section 5.3" or a file path). Locate the plan task spec and the drafted .tex file yourself.

Step 1: Read the Plan Spec

Find the plan file under docs/superpowers/plans/. Read the full task spec for this section. Extract:

  • The content checklist (every required item: definitions, theorems, propositions, examples, figures, exercises, cross-references).
  • The page budget (target pages and tolerance).
  • Required label names (e.g., \label{def:fpr}, \label{thm:composition}).
  • Required notation introductions.
  • Required forward and backward cross-references.

Step 2: Read the Drafted Section

Read the .tex file. Read the header comment block (DEFINED / RESOLVED / FORWARD lines). Note the approximate word count and estimated page length.

Step 3: Verify Each Checklist Item

For each item in the content checklist, determine whether it is present in the drafted section. When present, cite the line number or paragraph where it appears. When absent, mark FAIL with a description of what is missing.

Order matters: if the plan specifies that definition D must precede theorem T, and the draft has them reversed, mark FAIL for ordering.

Step 4: Check Page Budget

Estimate the compiled page length from word count and environment density. If the estimate falls outside the plan's tolerance, mark accordingly. Use the book's established ratio: approximately 400 words per page for mixed math/prose content.

Step 5: Check Label Definitions

For every label the plan requires, verify it appears in the draft's \label{} declarations. For every label in the header's DEFINED list, verify a matching \label{} exists in the body. Report mismatches.

Step 6: Check Cross-Reference Header Block

Verify that:

  • Every \label{} declared in the body appears in the DEFINED list.
  • Every \Cref{} or \ref{} to a prior section appears in RESOLVED.
  • Every \Cref{} or \ref{} to a future section appears in FORWARD.

Read the full file on GitHub · 84 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. 4d ago First seen · 84 lines · 44 tokens per session scan A beee577758a9

Subscribe to this mod's changes

spec-auditor is an agent published in the GitHub repository queelius/claude-anvil (2 stars, last pushed 1mo ago), licensed MIT. It adds 44 tokens to every session and 851 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

interview-prep

Use this agent for interview preparation, mock interviews, behavioral question coaching, technical interview practice, and company-specific research. Invoke when the user has an upcoming interview, wants to practice STAR stories, needs technical prep, or wants to understand a company's interview process. Context: User…

Aznatkoiny/zAI-Skills · 242 tokens

mentor

Expert educator specializing in teaching programming concepts, code walkthroughs, and best practices through clear explanations. Japanese: メンター.

Fujigo-Software/f5-framework-claude · 0 tokens

edtech-reviewer

Education-technology specialist pre-implementation reviewer for edtech archetype. Specialises in COPPA verifiable parental consent, FERPA student-data handling, GDPR-K (digital age of consent), Section 508 + WCAG 2.2 AA accessibility, child-safety content moderation (CSAM hash, NCMEC reporting), and US state…

avelikiy/great_cto · 112 tokens

ia-database-guardian

Reviews database schema, constraints, and migration code for safety. Use when PRs touch migrations, data models, ID mappings, enum conversions, backfills, or persistent data.

iliaal/whetstone · 41 tokens

ia-best-practices-researcher

Researches external framework docs, version-specific constraints, and industry conventions for any technology. Use when you need authoritative external documentation.

iliaal/whetstone · 34 tokens

reviewer-opus

Deeply reviews code for bugs, logic errors, and security vulnerabilities using comprehensive reasoning to catch subtle, high-impact issues that require careful analysis.

Joncik91/ucai · 32 tokens