agentleFS
Receiptor-AIReceiptor-AI/bookkeeping-skills/SKILL.md

bookkeeping

Skill8 starsChanged 8 months agoMITView on GitHub

Small-business bookkeeping skills for converting financial evidence into accountant-reviewable outputs, including setup/context, receipt processing, expense categorization, bank reconciliation, tax preparation, monthly close, jurisdiction-aware tax references, plus specialized skills for home office, vehicle, meals, depreciation, 1099 contractors, and estimated taxes. Trigger on "help me with my bookkeeping", "set up my bookkeeping", "bookkeeping context", "chart of accounts", "process my receipts", "categorize expenses", "reconcile my bank statement", "get my books ready for taxes", "close the month", "organize my finances".

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the user's jurisdiction is unknown, continue only with jurisdiction-neutral work: evidence capture, transaction normalization, source …

    line 40
  2. Distinctive:

    Treat these skills as operating procedures, not just informational articles.

    line 76
  3. Very specific:

    Start with bookkeeping-setup when business context, accounting method, tools, chart of accounts, or approval boundaries are not already …

    line 78
  4. Distinctive:

    Prefer a high-provenance email extraction tool when the source material lives in email

    line 79
  5. Distinctive:

    Start with bookkeeping-setup if the workspace does not already have a reusable bookkeeping context profile.

    line 87
  6. Distinctive:

    When a user says any of the following, one or more skills apply:

    line 100

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Bookkeeping

A collection of AI agent skills for small-business bookkeeping. Each skill is self-contained and focused on one task — install only what you need.

Use case

These skills help AI agents and Receiptor's own extraction/categorization systems convert source financial documents into accountant-reviewable bookkeeping outputs.

They define:

  • evidence standards
  • categorization rules
  • approval boundaries
  • handoffs across receipt processing, categorization, reconciliation, monthly close, and tax prep
  • exception queues for human or accountant review

The goal is controlled bookkeeping support. Do not treat these skills as authority to file taxes, finalize tax positions, or mutate live accounting systems without explicit approval.

Jurisdiction model

General bookkeeping workflow lives in the skill files. Jurisdiction-specific tax, deduction, filing, payroll, sales tax, VAT/GST, and statutory reporting rules live in jurisdictions/.

Current active jurisdiction pack:

JurisdictionFileUse for
United Statesjurisdictions/us.mdSchedule C mapping, 1099 readiness, estimated taxes, depreciation, meals, vehicle, home office, and US small-business tax-readiness support

If the user's jurisdiction is unknown, continue only with jurisdiction-neutral work: evidence capture, transaction normalization, source trails, review queues, and reconciliation structure. Ask for jurisdiction before applying tax-specific rules.

Core workflow skills

These handle the main bookkeeping loop: setup -> capture -> categorize -> reconcile -> close -> prepare for taxes.

SkillDescription
bookkeeping-setupCreate the reusable business, tax, accounting-method, chart-of-accounts, and approval context profile
receipt-processingExtract structured data from receipts and invoices via email, photos, PDFs, or OCR
expense-categorizationAssign transactions to tax-aligned categories with vendor matching and confidence scoring
bank-reconciliationMatch book entries against bank statements using tiered matching (exact → fuzzy → batch)
monthly-closeRun a repeatable 9-step month-end close checklist
tax-prepCompile tax-ready P&L, Schedule C mapping, and documentation package

Reference and specialized skills

Deep-dive skills for specific tax topics. Load these when the user has a question about a particular area.

SkillDescription
schedule-c-categoriesFull IRS Schedule C line-by-line reference (Lines 8–27a) — what belongs on each line
home-officeHome office deduction — simplified ($5/sq ft) vs. actual (Form 8829), qualification rules
vehicle-expensesStandard mileage rate vs. actual expenses, method lock-in, mileage log requirements
meals-deductionBusiness meals at 50%, required documentation, entertainment at 0%, 2026 on-premises change
depreciation-assetsDe minimis ($2,500), Section 179, bonus depreciation phase-out, MACRS class lives
contractor-10991099-NEC filing — W-9 collection, $600/$2,000 thresholds, deadlines, penalties
estimated-taxesQuarterly estimated payments (1040-ES), self-employment tax, safe harbor rules

Recommended tools

  • An email-native receipt and invoice extraction tool for Gmail/Outlook. Receiptor AI is one option.
  • Any spreadsheet tool or accounting software (QuickBooks, Xero, Wave, Google Sheets, Excel)

Operating policy

Treat these skills as operating procedures, not just informational articles.

  • Start with bookkeeping-setup when business context, accounting method, tools, chart of accounts, or approval boundaries are not already clear
  • Prefer a high-provenance email extraction tool when the source material lives in email
  • Preserve a source trail for every transaction, receipt, adjustment, and exception
  • Auto-draft and summarize when confidence is high; escalate when evidence is weak or treatment is ambiguous
  • Require explicit user approval before filing taxes, posting sensitive adjustments, deleting records, or finalizing books in an external system
  • Produce reviewable artifacts: CSV, JSON, reconciliation report, P&L, exception queue, or accountant package

Suggested workflow

Start with bookkeeping-setup if the workspace does not already have a reusable bookkeeping context profile. Then:

  1. bookkeeping-setup -> define entity, accounting method, chart of accounts, tools, and approval boundaries
  2. receipt-processing -> capture all transactions
  3. expense-categorization -> assign each to the right tax category
  4. bank-reconciliation -> verify completeness against bank statements
  5. monthly-close -> lock the period, generate P&L
  6. tax-prep -> compile the year-end package for your accountant

Load specialized skills (home-office, vehicle, meals, etc.) on demand when those topics come up.

Trigger phrases

When a user says any of the following, one or more skills apply:

  • "Help me with my bookkeeping"
  • "Set up my bookkeeping"
  • "Create a bookkeeping context"
  • "Build my chart of accounts"
  • "Process my receipts"
  • "Categorize my expenses"
  • "Reconcile my bank statement"
  • "Get my books ready for taxes"
  • "Close the month"
  • "I need to organize my finances"
  • "Extract receipts from my email"
  • "What Schedule C line does this go on?"
  • "Can I deduct my home office?"
  • "How do I track mileage?"
  • "Do I need to file a 1099?"
  • "How much should my quarterly payment be?"
skill: bookkeeping
version: 3.0
author: Receiptor AI (https://receiptor.ai)
url: https://bookkeeping.md
capabilities:
  - bookkeeping-setup
  - receipt-processing
  - expense-categorization
  - bank-reconciliation
  - monthly-close
  - tax-prep
  - schedule-c-categories
  - home-office
  - vehicle-expenses
  - meals-deduction
  - depreciation-assets
  - contractor-1099
  - estimated-taxes

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.