tax-classifier

tax-classifier is a skill for Claude Code from michielinksee/bantou. It costs 33 tokens per session (764 once invoked), scanned A, original, MIT.

A system for assigning Japanese business transactions to accounting categories, called 勘定科目, and consumption-tax treatments. It checks exclusions, past corrections, keywords, and then uses AI when needed.

In plain words
What is it for?
It helps classify transactions, assign tax treatment, and return a confidence score for accounting review.
Why use it?
It separates entries that should not be classified and uses remembered decisions before more costly AI processing.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions CLAUDE.md.

Part of the bantou plugin — 5 skills, 6 commands, 1 MCP server shipped together

Good fit It helps classify transactions, assign tax treatment, and return a confidence score for accounting review.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/michielinksee/bantou/tax-classifier
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.

Any agent
npx skills add michielinksee/bantou --skill tax-classifier
Clone the repo
git clone --depth 1 https://github.com/michielinksee/bantou

Made for: Claude Code.

Or install bantou, the plugin that ships this one along with the rest of its 5 skills, 6 commands, 1 MCP server.

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 tax-classifier

README.md
[![agentmods](https://agentmods.dev/badge/skills/michielinksee/bantou/tax-classifier/github.svg)](https://agentmods.dev/skills/michielinksee/bantou/tax-classifier)
Your own site
<a href="https://agentmods.dev/skills/michielinksee/bantou/tax-classifier"><img src="https://agentmods.dev/badge/skills/michielinksee/bantou/tax-classifier/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 tax-classifier

Your own site · 80×15
<a href="https://agentmods.dev/skills/michielinksee/bantou/tax-classifier"><img src="https://agentmods.dev/badge/skills/michielinksee/bantou/tax-classifier.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 33 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 764 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 $0.00033 $0.00764
Opus 5 $0.00016 $0.00382
Sonnet 5 $0.00007 $0.00153
Haiku 4.5 $0.00003 $0.00076

Measured 11d ago against content hash 1d2eb41e91b3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

tax-classifier 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 11d 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.

packages/claude-for-jp-accounting/skills/tax-classifier/SKILL.md · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Tax Classifier

Classify Japanese business transactions into standard accounting categories (kanjou-kamoku / 勘定科目) using a 3-stage pipeline that minimizes API costs and maximizes accuracy.

When to use

Invoke this skill whenever the user provides a transaction description (e.g. "スタバ 渋谷店 ¥580") and needs:

  • A 勘定科目 classification
  • A 税区分 (consumption-tax treatment) assignment
  • A confidence score

Pipeline overview

Stage 0 — Exclusion check (7 rules)

Before any classification attempt, run check_exclusion to filter out non-classifiable items. The standard rules exclude:

  1. Transfer-to-self (振替) entries
  2. Opening / closing balance lines
  3. Duplicate reversed entries (赤伝)
  4. Summary / subtotal rows
  5. Tax-payment lines handled separately
  6. Payroll deduction detail lines
  7. Memo-only entries with zero amount

If an item is excluded, return the exclusion reason without proceeding.

Stage 1 — Memory recall + keyword dictionary (14 categories)

Query recall_memory first. If the exact vendor or pattern has been corrected before, use the memorized classification immediately.

If no memory hit, match against the 14-category keyword dictionary:

Category Examples
会議費 cafe, meeting room
交際費 dinner with client, gift
旅費交通費 train, taxi, hotel
通信費 phone, internet, postage
消耗品費 office supplies, toner
地代家賃 rent, parking
水道光熱費 electricity, gas, water
支払手数料 bank fee, service charge
広告宣伝費 ads, marketing
外注費 subcontractor, freelancer
福利厚生費 employee meal, health check
保険料 insurance premium
租税公課 stamp duty, property tax
雑費 miscellaneous (last resort)

Stage 2 — Claude AI fallback

When keyword matching fails or confidence is below threshold, call classify_transaction which uses Claude to analyze the description with full context from the firm's CLAUDE.md configuration.

Consumption tax handling

Always assign a tax treatment. Be careful with the rates:

  • 10% — standard rate (標準税率)
  • 8% reduced — food/beverage for takeout, newspaper subscriptions (軽減税率)
  • Exempt (非課税) — land rent, insurance, interest
  • Non-taxable (不課税) — donations, damages, foreign transactions

Read the full file on GitHub · 93 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. 11d ago First seen · 93 lines · 33 tokens per session scan A 1d2eb41e91b3

Subscribe to this mod's changes

tax-classifier is a skill published in the GitHub repository michielinksee/bantou (2 stars, last pushed 19d ago), licensed MIT. It adds 33 tokens to every session and 764 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 skills, from other repositories

bank-reconciliation

Match book entries against bank or credit card statements to find discrepancies, missing transactions, and duplicates. Uses exact, near, and batch matching with balance verification. Trigger on "reconcile bank statement", "match transactions", "bank balance doesn't match", "close the books".

Receiptor-AI/bookkeeping-skills · 58 tokens

monthly-close

Repeatable month-end close checklist for small businesses — lock the period, process outstanding receipts, categorize remaining items, reconcile all accounts, review large transactions, check contractor 1099 thresholds, generate monthly P&L summary. Trigger on "close the month", "month-end close", "monthly…

Receiptor-AI/bookkeeping-skills · 68 tokens

depreciation-assets

Determine whether to expense or depreciate business assets using the de minimis safe harbor ($2,500), Section 179 ($1,220,000 limit), bonus depreciation (phase-out 2024–2027), or MACRS schedules. Track asset basis, useful life, and accumulated depreciation. Trigger on "depreciate", "Section 179", "bonus depreciation"…

Receiptor-AI/bookkeeping-skills · 95 tokens

estimated-taxes

Calculate and schedule quarterly estimated tax payments (Form 1040-ES) for self-employed individuals. Covers self-employment tax calculation (15.3%), income tax estimation, safe harbor rules (100%/110% of prior year), payment deadlines, and underpayment penalties. Trigger on "estimated taxes", "quarterly taxes"…

Receiptor-AI/bookkeeping-skills · 96 tokens

meals-deduction

Rules for deducting business meals at 50%, required documentation (who, what, where, why), entertainment at 0% since TCJA 2018, on-premises meals going to 0% in 2026, and common mistakes. Trigger on "deduct a meal", "business lunch", "client dinner deductible", "meals deduction", "entertainment deduction", "can I…

Receiptor-AI/bookkeeping-skills · 93 tokens

tax-prep

Organize financial records into tax-ready reports — P&L mapped to Schedule C lines, quarterly breakdowns, 1099 contractor summary, and flagged items requiring special treatment. Does not file taxes or give tax advice. Coordinates with specialized skills for meals, depreciation, home office, vehicle, and 1099 details.…

Receiptor-AI/bookkeeping-skills · 94 tokens