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.
npx skills add Receiptor-AI/bookkeeping-skills --skill expense-categorizationgit clone --depth 1 https://github.com/Receiptor-AI/bookkeeping-skillsWrote 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/receiptor-ai/bookkeeping-skills/expense-categorization)<a href="https://agentmods.dev/skills/receiptor-ai/bookkeeping-skills/expense-categorization"><img src="https://agentmods.dev/badge/skills/receiptor-ai/bookkeeping-skills/expense-categorization/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/receiptor-ai/bookkeeping-skills/expense-categorization"><img src="https://agentmods.dev/badge/skills/receiptor-ai/bookkeeping-skills/expense-categorization.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 | $0.00071 | $0.01992 |
| Opus 5 | $0.00036 | $0.00996 |
| Sonnet 5 | $0.00014 | $0.00398 |
| Haiku 4.5 | $0.00007 | $0.00199 |
Grade A, and why
expense-categorization 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 10d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Expense Categorization
Assign every transaction to the correct expense category so it flows to the right line on a tax return, produces accurate financial statements, and survives an audit.
Read these when needed
- Read references/DECISION-RULES.md when deciding whether a category is safe to auto-assign or should go to review.
- Run
scripts/category_review_summary.pywhen you already have categorized records and need a deterministic summary by confidence, category, and review status.
Why categorization matters
The IRS doesn't audit "total expenses." It audits individual line items on Schedule C (or the equivalent form for your entity type). A business that lumps everything into "Other Expenses" invites scrutiny. Proper categorization also reveals spending patterns — you can't cut costs you can't see.
For the full Schedule C line-by-line reference (Lines 8–27a with what belongs and what doesn't), see the schedule-c-categories skill.
Categorization logic
Primary signal: vendor name
Most categorization comes from the vendor name alone. Use the mapping table below to auto-categorize high-confidence vendors.
Secondary signal: line items
When the vendor is ambiguous (Amazon, Walmart, Costco), look at line items from the receipt. This is where high-quality extraction is especially useful — line items matter more than totals. Amazon with "USB-C Hub, Monitor Stand" → Equipment. Amazon with "Paper Towels, Hand Soap" → Office Supplies. Mixed orders should be split across categories.
Tertiary signal: amount patterns
Recurring monthly charges at the same amount → likely Software & Subscriptions (Line 27a). Round-number transfers to individuals → likely Contractors (Line 11). Small charges at restaurants → Meals (Line 24b).
Learning from corrections
When the user corrects a categorization, store that mapping. User-confirmed mappings should override default logic going forward.
High-confidence vendor mapping (auto-categorize)
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 171 lines · 71 tokens per session scan A dcd96a022990
expense-categorization is a skill published in the GitHub repository Receiptor-AI/bookkeeping-skills (4 stars, last pushed 3mo ago), licensed MIT. It adds 71 tokens to every session and 1,992 once invoked, about $0.0004 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
correction-memory
A local memory system that stores tax-accountant corrections to transaction classifications. It matches future transactions to earlier vendor or description patterns.
nightly-batch
Batch-process all client companies overnight. Fetches unprocessed transactions from freee API, runs the full classification pipeline, and generates a summary for morning review.
tax-classifier
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.
csv-importer
A tool for importing transaction CSV files—spreadsheet files separated by commas—from Japanese accounting software and classifying every imported row.
monthly-closer
A monthly accounting report generator that summarizes transactions by account category and compares them with the previous month.
tax-calculator
Calculate tax implications of trades and generate tax reports.