For AI agents: Markdown version of this page · llms.txt
Setup guide

Transaction Categorization Agent for Town

The $19 Transaction Categorization Agent runs on Town as three plain-text files: set it up once in Town, approve a sample-data test, then trigger each run with the usage prompt and this run's inputs.

Get the Transaction Categorization Agent - $19Town overview

What the Transaction Categorization Agent does

Town already runs your inbox and calendar - this agent becomes one more job it owns. It proposes a category for every uncoded transaction with the evidence: date, payee, amount, proposed category, evidence, review status. You approve the batch.

Uncoded transactions pile up all month, and categorizing them means reading payee strings and trying to remember how you coded the same vendor last time.

Where the three files go in Town

Share the three files with Town and ask it to keep skill.md as the standing rules for the job, then run the setup prompt and approve the fictional-data test. Town's email and calendar focus pairs well with agents whose output is a draft, a schedule or a follow-up queue.

What a run looks like on Town

A run looks like this: you ask Town to run transaction-categorization with this run's inputs, and it works through the standing rules you gave it. The output - it proposes a category for every uncoded transaction with the evidence: date, payee, amount, proposed category, evidence, review status - is prepared for your review alongside the rest of your delegated work.

Setup, step by step

  1. Share skill.md, setup-prompt.txt and usage-prompt.txt with Town and ask it to keep the skill as the standing rules for the job (for example: transaction categorization).
  2. Run the setup prompt; the agent runs transaction-categorization on fictional sample data first - approve that test before real work.
  3. Town's inbox-and-calendar focus pairs best with agents whose output is a draft, a schedule or a follow-up queue.
  4. Trigger each real run with one message and this run's inputs.

Honest limits on Town

Town is inbox-and-calendar centric and its feature set is young; confirm its current file and memory handling in Town's own documentation before a big setup.

Questions people ask

Does the Transaction Categorization Agent work on Town?

Yes - the agent is three plain-text files, and on Town they live in Town. The four setup steps below are the whole job: Share skill.md, setup-prompt.txt and usage-prompt.txt with Town and ask it to keep the skill as the standing rules for the job (for example: transaction categorization). Then run the setup prompt; the agent runs transaction-categorization on fictional sample data first - approve that test before real work.

What does the Transaction Categorization Agent produce on Town?

It proposes a category for every uncoded transaction with the evidence: date, payee, amount, proposed category, evidence, review status. Town prepares it as part of your delegated queue; you approve before use. Every run waits for your approval before anything is sent, spent, published or scheduled.

Is this affiliated with Town?

No. AI Agent Skills is an independent store. We are not affiliated with, endorsed by, or sponsored by Town or its makers. Platform names belong to their owners; capabilities change, so check Town's current documentation.

Run this agent on another assistant

Instinct Claude Gemini Hermes OpenClaw ChatGPT Grok DeepSeek Perplexity Microsoft Copilot Muse

More bookkeepers agents on Town

Reconciliation Agent Receivables Follow Up Agent Bills Invoices Agent

AI Agent Skills is an independent store, not affiliated with Town or its makers. Platform names belong to their owners.