Transaction Categorizer

Assigns a category and subcategory to a financial transaction by matching its raw description against a configurable taxonomy and rules table, falling back to LLM inference when no rule matches. Emits a normalized merchant name, category path, recurring flag, and confidence score, and proposes new rules from confirmed classifications. Use when categorizing bank, credit-card, or brokerage transactions, building or refining a category taxonomy, or when user mentions transaction categorization, merchant normalization, expense classification, or category rules.

200 stars 25 forksFinance & Crypto Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add lyndonkl/claude

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add lyndonkl/claude --skill transaction-categorizer
Manual install
git clone --depth 1 https://github.com/lyndonkl/claude.git /tmp/claude
mkdir -p ~/.claude/skills
cp -r /tmp/claude/skills/transaction-categorizer ~/.claude/skills/transaction-categorizer

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Transaction Categorizer

## Table of Contents
- [Overview](#overview)
- [Input contract](#input-contract)
- [Workflow](#workflow)
- [Taxonomy](#taxonomy)
- [Rule format](#rule-format)
- [Confidence and audit](#confidence-and-audit)
- [Output contract](#output-contract)
- [Guardrails](#guardrails)

## Overview

A transaction is just a `description_raw` string and a signed `amount_cents`. This skill turns that into a clean `merchant`, a `category` + `subcategory`, an `is_recurring` boolean candidate, and a confidence. It applies rules first (cheap, deterministic) and only falls back to LLM inference for the residual.

It also produces *learned rules* — when a confident classification matches a clear merchant pattern, propose a new rule for the rule table so future identical transactions match for free.

## Input contract

The caller provides:

- `transactions` — array of `{id, description_raw, amount_cents, account_id, date, account_type}`.
- `taxonomy` — the `categories.json` taxonomy block (top-level → subcategory list).
- `rules` — array of existing rules (see [Rule format](#rule-format)).
- `account_type_hints` (optional) — when known, helps disambiguate (e.g., a deposit on a brokerage account is more likely `dividends` than `salary`).

## Workflow
Read the full skill on GitHub

Details

Skill name
transaction-categorizer
Path
skills/transaction-categorizer/SKILL.md
Repo updated
1 September 2026

FAQ

What does the Transaction Categorizer skill do?

Assigns a category and subcategory to a financial transaction by matching its raw description against a configurable taxonomy and rules table, falling back to LLM inference when no rule matches. Emits a normalized merchant name, category path, recurring flag, and confidence score, and proposes new rules from confirmed classifications. Use when categorizing bank, credit-card, or brokerage transactions, building or refining a category taxonomy, or when user mentions transaction categorization, merchant normalization, expense classification, or category rules.

How do I install the Transaction Categorizer skill?

Claude Code plugin: /plugin marketplace add lyndonkl/claude. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Transaction Categorizer popular?

It lives in lyndonkl/claude, which has 200 GitHub stars and 25 forks.

Related skills

Abstraction Concrete Examples

200

lyndonkl/claude

Builds structured abstraction ladders that translate high-level principles into concrete, actionable examples across 3-5 levels. Bridges communication gaps, reveals hidden assumptions, and tests whether abstract ideas work in practice. Use when explaining concepts at different expertise levels, moving between abstract principles and concrete implementation, identifying edge cases by testing ideas against scenarios, designing layered documentation, decomposing complex problems into actionable steps, or bridging strategy-execution gaps.

Testing & QA25

Academic Letter Architect

200

lyndonkl/claude

Guides the creation of evidence-based academic recommendation letters, reference letters, and award nominations that combine concrete examples, meaningful comparisons, and genuine enthusiasm. Use when writing recommendation letters for students, postdocs, or colleagues, or when user mentions recommendation letter, reference, nomination, letter of support, endorsement, or needs help with strong advocacy and comparative statements.

Research & Science25

Adr Architecture

200

lyndonkl/claude

Documents significant architectural and technical decisions with full context, alternatives considered, trade-offs analyzed, and consequences understood. Creates a decision trail that helps teams understand why decisions were made. Use when choosing between technology options, making infrastructure decisions, establishing standards, migrating systems, or when user mentions ADR, architecture decision, technical decision record, or decision documentation.

Software Development25

Transaction Deduplicator

200

lyndonkl/claude

Detects and removes duplicate transactions across overlapping bank, credit-card, and brokerage statement imports using a stable composite key (account_id, date ±1d, amount_cents, description_normalized). Emits a list of new transactions to commit, a list of suppressed duplicates with their reasons, and a list of suspicious near-duplicates that need human review. Use when ingesting financial statements that may overlap prior drops, merging multiple export sources for the same account, or when user mentions duplicate transactions, deduping a transaction file, or reconciling overlapping statement

Finance & Crypto25

Tax Loss Harvest Scanner

200

lyndonkl/claude

Scans a taxable brokerage account for individual lots with unrealized losses above a configurable threshold, identifies wash-sale risks by checking recent buys and forward planned buys of substantially identical securities (across all household accounts including spousal), and proposes harvest pairs (sell-for-loss + immediate buy of a similar-but-not-identical replacement). Use for year-end tax planning, monthly TLH scans, after market drawdowns, or when user mentions tax-loss harvesting, TLH, wash sale, or harvest candidates.

Finance & Crypto25

Valuation Reporting

200

lyndonkl/claude

Assembles a finished company analysis into the deliverable a decision-maker reads — the executive summary, the value bridge, the assumptions that matter, the sensitivity range, and the recommendation with its catalysts and risks. Use when writing up a valuation, DCF, acquisition, IPO, restructuring or corporate finance analysis, when drafting REPORT.md or the verdict, or when deciding what belongs in the report.

Finance & Crypto25

Browse the most popular Claude skills, all finance & crypto skills, or write your own with the skills generator.