Setup guides

AI agents for bookkeepers on Instinct

The work is accuracy at volume: uncoded transactions, unmatched rows, invoices nobody chased. These four agents prepare the review queues with evidence. Posting and sending stay with you.

The four agents, on Instinct

Transaction Categorization Agent for Instinct

It proposes a category for every uncoded transaction with the evidence: date, payee, amount, proposed category, evidence, review status.

Setup guide →

Reconciliation Agent for Instinct

It matches statement rows to ledger rows and hands you a complete exception list: match status, difference and next action per row.

Setup guide →

Receivables Follow Up Agent for Instinct

It tracks open invoices and prepares reminder drafts: customer, invoice, amount due, days overdue, reminder draft and approval status.

Setup guide →

Bills Invoices Agent for Instinct

It captures new bills and invoices, runs duplicate checks and builds a due-item review queue: vendor, invoice, amount, due date, duplicate check, review status.

Setup guide →

How setup works on Instinct

Hand the three files (skill.md, setup-prompt.txt, usage-prompt.txt) to Instinct in chat. It keeps the context across sessions, so the setup survives: run the setup prompt once, approve the fictional-data test, and the agent becomes a standing workflow you trigger with one line. This is the closest fit to how the agents are designed to run.

Honest limits

Instinct is in early access and its capabilities are expanding; always approve the sample-data test before pointing the agent at real work.

Questions people ask

Which AI agents for bookkeepers run on Instinct?

All four: Transaction Categorization Agent, Reconciliation Agent, Receivables Follow Up Agent, Bills Invoices Agent. Each is $19 one-time - three plain-text files that live in Instinct.

Why run bookkeepers agents on Instinct?

Hand the three files (skill.md, setup-prompt.txt, usage-prompt.txt) to Instinct in chat. It keeps the context across sessions, so the setup survives: run the setup prompt once, approve the fictional-data test, and the agent becomes a standing workflow you trigger with one line. This is the closest fit to how the agents are designed to run.

Is this affiliated with Instinct?

No. AI Agent Skills is an independent store, not affiliated with or endorsed by Instinct or its makers.

AI Agent Skills is an independent store, not affiliated with Instinct or its makers.