Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/michielinksee/bantounpx agentmods add commands/michielinksee/bantou/importWrote 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.
[](https://agentmods.dev/commands/michielinksee/bantou/import)<a href="https://agentmods.dev/commands/michielinksee/bantou/import"><img src="https://agentmods.dev/badge/commands/michielinksee/bantou/import/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.
<a href="https://agentmods.dev/commands/michielinksee/bantou/import"><img src="https://agentmods.dev/badge/commands/michielinksee/bantou/import.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00014 | $0.00476 |
| Opus 5.5 | $0.00006 | $0.00190 |
| Sonnet 5.5 | $0.00003 | $0.00095 |
| Haiku 4.5 | $0.00001 | $0.00048 |
Grade A, and why
import 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 1mo 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.
What it actually says
/bantou:import
Import a CSV file from Japanese accounting software and classify all transactions automatically.
Usage
/jp-accounting:import <file-path>
Examples
/jp-accounting:import ./exports/yayoi_202601.csv
/jp-accounting:import C:\Users\会計\freee_export.csv
/jp-accounting:import ~/Downloads/mf_journal.csv
Input
An absolute or relative path to a CSV file. The importer auto-detects:
- File format — Yayoi (Format A/B), freee, MoneyForward, or generic
- Encoding — UTF-8, Shift_JIS, CP932 (auto-converted to UTF-8)
- Delimiter — comma, tab, or semicolon
Processing flow
- Parse the CSV file and detect format
- Validate headers and data structure
- Run each row through the classification pipeline:
- Stage 0: Exclusion filter
- Stage 1: Memory recall + keyword matching
- Stage 2: AI fallback (for unmatched rows)
- Generate a summary report
Output
After processing, a summary table is displayed:
インポート結果: yayoi_202601.csv
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
総行数: 142
除外 (Stage 0): 12
自動分類 (高信頼): 98
自動分類 (中信頼): 18
要確認 (低信頼): 14
Memory ヒット: 34 (26.2%)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Low-confidence items are listed individually for review.
Encoding troubleshooting
If characters appear garbled (文字化け), try:
- Open the CSV in a text editor and re-save as UTF-8
- Specify encoding explicitly when prompted
- Check that the file was not corrupted during download
Notes
- Large files (>1,000 rows) are processed in batches of 100
- Progress is shown during processing
- Results can be exported or used with
/jp-accounting:report
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.
- 1mo ago First seen · 73 lines · 14 tokens per session scan A b2aee7d7165d
import is a command published in the GitHub repository michielinksee/bantou (2 stars, last pushed 2mo ago), licensed MIT. It adds 14 tokens to every session and 476 once invoked, about $0.0001 per session on Opus 5.5. 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.
Other commands, from other repositories
report
Generate a financial report from the book — a styled HTML page and/or an Excel workbook (balance sheet, income statement, transactions).
jaz-fa-review
Review the fixed asset register in Jaz: check depreciation, identify disposals and write-offs needed.
jaz-tax-sg
Run Singapore Form C-S / C-S Lite corporate income tax computation using Jaz data: pull P&L, classify add-backs, compute capital allowances, apply exemptions.
jaz-audit-prep
Compile an audit preparation pack in Jaz: generate reports, schedules, and reconciliations for auditor or tax agent.
jaz-gst
Prepare GST/VAT filing in Jaz: review tax ledger, identify discrepancies, generate filing summary.
jaz-recon
Run bank reconciliation in Jaz: match bank records to transactions, categorize unmatched items, resolve discrepancies.