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/cynco-labs/ai-accounting-skillsnpx agentmods add skills/cynco-labs/ai-accounting-skills/classify-transactionsWrote 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/skills/cynco-labs/ai-accounting-skills/classify-transactions)<a href="https://agentmods.dev/skills/cynco-labs/ai-accounting-skills/classify-transactions"><img src="https://agentmods.dev/badge/skills/cynco-labs/ai-accounting-skills/classify-transactions/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/skills/cynco-labs/ai-accounting-skills/classify-transactions"><img src="https://agentmods.dev/badge/skills/cynco-labs/ai-accounting-skills/classify-transactions.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 | $0.00055 | $0.01372 |
| Opus 5 | $0.00028 | $0.00686 |
| Sonnet 5 | $0.00011 | $0.00274 |
| Haiku 4.5 | $0.00006 | $0.00137 |
Grade A, and why
classify-transactions 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 12d 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.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/classify-transactions
Purpose
Codes are the end of classify. Substance is the middle.
Main job: classify.
Doctrine: shared/classify-substance.md.
Script first for baseline; agent applies economic substance for material themes when depth requires it.
Preconditions
shared/guardrails.md+ firm profile if present.- Active engagement;
workpapers/transactions.jsonexists (extract done). - Framework known or provisional (
engagement_state.json). - Re-read sources if context was compacted (files are the books).
- Load
shared/classify-substance.mdand chooseclassify_depth:bookkeeping— fast pathstandards_aware— year-end / “proper classifications” / revenue or capex questions
Write classify_depth into state (notes or artifacts map) when set.
Steps
1 — Run the classifier (required)
python3 scripts/classify_transactions.py \
--input workpapers/transactions.json \
--output workpapers/transactions.json \
--payee-map workpapers/payee_map.json \
--report workpapers/classification_review.md
# or
npx @cynco/accounting-skills classify ./workpapers/transactions.json
Rules: references/classification_patterns.json.
Payee map (optional): workpapers/payee_map.json.
| Field | Meaning |
|---|---|
classification_basis |
prior_year · pattern · invoice · employee · suspense · analysis |
classification_confidence |
0–1 |
needs_review |
Agent must confirm |
Do not re-derive pattern tables in chat when the script exists.
Done when: script exit 0 and classification_review.md written.
2 — Money map (material themes)
From transactions.json + review report, group material lines:
| Theme | Signal | Checklist (Malaysia pack) | Analysis file |
|---|---|---|---|
| Revenue | Material credits / customer-like inflows | references/jurisdictions/malaysia/standards/revenue_recognition.md |
workpapers/analysis/revenue_recognition.md |
| Capital vs expense | Material one-off / asset-like outflows | …/standards/capital_vs_expense.md |
workpapers/analysis/capital_vs_expense.md |
| Related parties | Directors / shareholders / common control | references/mpers.md S33 |
workpapers/analysis/related_parties.md |
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.
- 12d ago First seen · 145 lines · 55 tokens per session scan A f45bd3ea871c
classify-transactions is a skill published in the GitHub repository cynco-labs/ai-accounting-skills (3 stars, last pushed 2mo ago), licensed MIT. It adds 55 tokens to every session and 1,372 once invoked, about $0.0003 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.
Other skills, from other repositories
countbean-accounting
Rules for writing correct double-entry Beancount transactions when managing a Countbean cloud book — account naming, how to balance postings, categorisation, and safe ingestion. Use whenever recording, importing, or correcting financial data in a Countbean book.
bookkeeping
Use when a small business needs audit-ready books — a chart of accounts, posting a transaction to the right account and side, clearing an uncategorized bank feed, cash vs accrual, or a ledger that won't tie to the bank. NOT interpreting the numbers — runway, burn, P&L cadence (that is finance-ops), NOT issuing…
jangbu-jongso
A Korean tax-document checklist generator for individuals, freelancers, sole-company owners, mixed earners, and certain business owners.
korean-jangbu-for
A Korean bookkeeping assistant for startups, freelancers, sole proprietors, and one-person companies. It processes financial files locally, helps classify transactions, and prepares accounting and tax outputs.
jangbu-import
A data-import workflow for turning bank files, card records, spreadsheets, receipts, tax invoices, and statement PDFs into a standard set of 13 transaction fields. OCR, or optical character recognition, is used to read information from document images and PDFs.
jangbu-connect
A setup guide for connecting Korean tax, bank, card, and social-insurance data through CODEF, an external API service. It explains how users obtain their own CODEF credentials and store them locally.