Bruin CLI Step 4 of 7

Build an AI Finance Agent for QuickBooks

1) See the context the template already ships

The template was written to be read by an agent, so most of the context layer exists before you write anything:

  • Column descriptions and checks - every asset declares its full schema with types, descriptions, and checks. On BigQuery, metadata_push also publishes the descriptions to the dataset.
  • meta on every report - the grain, the questions the table answers, whether it holds personal data, and agent_notes such as "work the queue newest first and read each description before acting"
  • quickbooks/AGENTS.md - the agent guide: which table answers which question, how MRR, net burn, runway, and DSO are defined, and three workflows (month-end review, categorizing transactions, and duplicates and anomalies)

Generate a browsable version of the same definitions to see what the agent sees:

bruin docs quickbooks -o quickbooks-docs.html
open quickbooks-docs.html

Sign up to our newsletter

Practical updates on open-source data pipelines, AI analysts, governance, and what we are shipping at Bruin.

The signup form is hosted by Brevo. Allow marketing cookies to load it.