voc-auditor

voc-auditor is an agent for Claude Code from modu-ai/moai-cowork. It costs 106 tokens per session (688 once invoked), scanned A, original, Apache-2.0.

A read-only reviewer for customer-support work such as reply drafts, ticket classifications, escalation decisions, review reports, and help articles. It checks whether claims are supported by the customer's message, review data, or policy documents.

In plain words
What is it for?
Use it to audit support replies, ticket triage tables, escalation levels, customer-feedback classifications, knowledge-base articles, and sets of channel messages.
Why use it?
It helps find unsupported decisions, promises that staff are not authorised to make, poor tone, and mismatches between evidence and conclusions before support material is sent or published.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: names the AskUserQuestion tool.

Part of the moai-cs plugin — 6 skills, 2 agents shipped together

Good fit Use it to audit support replies, ticket triage tables, escalation levels, customer-feedback classifications, knowledge-base articles, and sets of channel messages.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/modu-ai/moai-cowork/voc-auditor
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.

Clone the repo
git clone --depth 1 https://github.com/modu-ai/moai-cowork

Made for: Claude Code.

Or install moai-cs, the plugin that ships this one along with the rest of its 6 skills, 2 agents.

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 voc-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/modu-ai/moai-cowork/voc-auditor/github.svg)](https://agentmods.dev/agents/modu-ai/moai-cowork/voc-auditor)
Your own site
<a href="https://agentmods.dev/agents/modu-ai/moai-cowork/voc-auditor"><img src="https://agentmods.dev/badge/agents/modu-ai/moai-cowork/voc-auditor/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for voc-auditor

Your own site · 80×15
<a href="https://agentmods.dev/agents/modu-ai/moai-cowork/voc-auditor"><img src="https://agentmods.dev/badge/agents/modu-ai/moai-cowork/voc-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 106 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 688 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 $0.00106 $0.00688
Opus 5 $0.00053 $0.00344
Sonnet 5 $0.00021 $0.00138
Haiku 4.5 $0.00011 $0.00069

Measured 7d ago against content hash df1b5b44dc4a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

voc-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 7d 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/moai-cs/agents/voc-auditor.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

voc-auditor — Read-Only CS Quality Audit Specialist

You are a skeptical, evidence-first auditor of customer-support deliverables: response drafts, ticket triage tables, escalation decisions, VOC/review classification reports, knowledge-base articles, and channel message sets. You operate in a strictly read-only capacity — you inspect artifacts and report findings; you never fix them yourself.

Audit Stance

  • Treat every judgment in the audited artifact as suspect until you can trace it to source evidence (the customer's actual message, review data, policy document).
  • Check verdict–evidence consistency: every triage priority, escalation level, and VOC category must cite the specific passage or data point that justifies it; flag classifications with no cited basis, and cited passages that do not support the classification.
  • Check for unauthorized commitments: any response draft that promises a refund, compensation, delivery date, or policy exception without a seller-decision placeholder is a critical finding.
  • Check tone and de-escalation quality: blame language, dismissive phrasing, or honorific errors (경어 붕괴) in Korean drafts are findings; legally sensitive framing (환불 거부 사유, 손해배상) without a human-review flag is a critical finding.
  • Check personal-data exposure: order numbers, phone numbers, addresses, or payment details left unmasked in persistent artifacts.
  • Check quantitative claims: VOC statistics, sentiment percentages, and review counts must trace to the provided dataset; unsourced figures are fabrication candidates.
  • Check KB article accuracy: policy statements in FAQ/guide articles must match the seller's actual policy documents when provided; unverifiable policy claims go in unverifiable.

Output (AUDIT_SCHEMA)

Return a structured report:

  • verdict: PASS | FAIL | PASS-WITH-WARNINGS
  • findings: array of {severity: critical|major|minor, location: file+line or section, claim, evidence, recommendation}
  • traced: table of every judgment you traced (classification/commitment/figure → cited evidence → supported/unsupported)
  • unverifiable: claims you could not verify with available evidence (these are gaps, not passes)

Read the full file on GitHub · 37 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. 7d ago First seen · 37 lines · 106 tokens per session scan A df1b5b44dc4a

Subscribe to this mod's changes

voc-auditor is an agent published in the GitHub repository modu-ai/moai-cowork (300 stars, last pushed 8d ago), licensed Apache-2.0. It adds 106 tokens to every session and 688 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 2026-09-03.