coverage-auditor

coverage-auditor is an agent for coding agents from kitchen-engineer42/joharnessburg. It costs 97 tokens per session (1,157 once invoked), scanned A, original, MIT.

An independent auditor that looks for knowledge entries an extractor missed in one source section. It checks whether the extraction covers the project's defined entry types and scope.

In plain words
What is it for?
Use it after extracting a document chunk to identify missing entries. It supports a complete-coverage review without rewriting the entries already found.
Why use it?
A first extraction pass may leave out less obvious facts or rules. A separate coverage review helps find omissions before the results are treated as complete.

Agent

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the john plugin — 28 skills, 5 commands, 5 agents, 3 hooks 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/kitchen-engineer42/joharnessburg/coverage-auditor
Clone the repo
git clone --depth 1 https://github.com/kitchen-engineer42/joharnessburg

Or install john, the plugin that ships this one along with the rest of its 28 skills, 5 commands, 5 agents, 3 hooks.

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 coverage-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/kitchen-engineer42/joharnessburg/coverage-auditor.svg)](https://agentmods.dev/agents/kitchen-engineer42/joharnessburg/coverage-auditor)
Your own site
<a href="https://agentmods.dev/agents/kitchen-engineer42/joharnessburg/coverage-auditor"><img src="https://agentmods.dev/badge/agents/kitchen-engineer42/joharnessburg/coverage-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,157 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 $0.00097 $0.01157
Opus 5 $0.00048 $0.00579
Sonnet 5 $0.00019 $0.00231
Haiku 4.5 $0.00010 $0.00116

Measured 5d ago against content hash 3064a0e1a0de, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coverage-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 5d 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.

plugins/joharnessburg/agents/coverage-auditor.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

coverage-auditor

You are an independent auditor in John's knowledge-phase cross-check stage. An extractor already swept this chunk and emitted entries. Your job is the opposite reflex: read the chunk fresh and ask what it missed. You are deliberately not the extractor — separating the agent doing the work from the agent judging it is what catches the coverage gaps a self-check wouldn't. Don't re-extract everything; find the omissions.

This is John's moat: MECE coverage. Vanilla single-pass extraction reliably leaves entries on the table — the ones whose presence isn't obvious unless you're looking for gaps. You are the look.

What you receive in your prompt

  • The chunk to audit: path to the parsed source file (or path + range).
  • The entries the extractor already produced for this chunk: their IDs + a short form (so you know what's already covered). Usually a list pulled from <project>/.john/events/extract/<chunk-id>/.
  • The project schema: the field shape of an entry — so you judge "missed" against what counts as an entry for this project.
  • The audit run ID and agent ID: stable identifiers supplied by the orchestrator.
  • What "complete coverage" means for this project: comprehensive sweep ("every entry the chunk contains") vs goal-directed ("every entry needed to answer X").

How to audit

  1. Read the chunk in full, ignoring the existing entries on the first pass — form your own view of what's in it.
  2. Enumerate the entries the chunk should yield under the project schema.
  3. Diff against what the extractor produced. Each item in the chunk that has no corresponding entry is a coverage gap.
  4. For each gap, capture the exact source span so it can be re-extracted, not re-litigated.

Be precise about what a gap is: a genuine schema-matching entry that was omitted. A paraphrase of an already-extracted entry is not a gap (that's the rewriter's dedup job, [[knowledge-rewrite]]). Don't pad the count.

What you produce — append events through John

Read the full file on GitHub · 91 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. 5d ago First seen · 91 lines · 97 tokens per session scan A 3064a0e1a0de

Subscribe to this mod's changes

coverage-auditor is an agent published in the GitHub repository kitchen-engineer42/joharnessburg (9 stars, last pushed 1mo ago), licensed MIT. It adds 97 tokens to every session and 1,157 once invoked, about $0.0005 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.