Clean up intake
Drop receipts, invoices, PDFs, and statements into the Book. The Agent creates drafts, preserves attachments, and lets the reviewer decide what gets posted.
Use Bkper with the AI assistant you already have
Ask ChatGPT or Claude to work with your books.
Ask questions, create reports and PDFs, review transactions, or build workflows. Bkper keeps the records, balances, permissions, and review history intact.
For accountants, owners, and operators navigating AI
Build your own workflows, not demo magic.
Reduce prep, surface review issues earlier, and make recurring work repeatable.
Review the book
Ask for unchecked transactions, unusual movements, missing support, duplicate candidates, or category patterns before month-end work gets expensive.
Build workpapers
Use ChatGPT Work or Claude to turn Bkper data into PDFs, scripts, and repeatable reports.
Modernize your accounting stack
Build tools, data connectors, and book-aware apps around your structure, permissions, and review process.
Why Bkper fits AI work
The book stays the source of truth.
Models can draft and reason, but financial facts should come from Bkper records and balances. Review the resulting drafts, scripts, reports, and Book activity.
Zero-sum records
Every posted transaction moves resources from one Account to another.
Draft before impact
Incomplete or AI-created work can remain draft until reviewed.
Deterministic balances
Reports come from Book records and APIs, not model arithmetic.
Visible agency
People, apps, bots, and agents are identified in the activity trail.
Scoped permissions
OAuth, Book roles, lock dates, and checked rules constrain what can happen.
Evidence attached
Files stay with the transactions they support.
Choose your starting point
Three clear ways to work with AI.
Choose the workflow that matches the job—not a model or protocol.
Use ChatGPT or Claude
Connect Bkper to your current assistant on the web or desktop. It keeps using its own models.
Set up your assistant →Use Bkper CLI Agent
Use one terminal workspace for Bkper context, local files, commands, scripts, tests, and included AI models.
Explore Bkper CLI Agent →Let Bkper Agent create drafts
Turn incoming documents into draft transactions without starting a chat.
Process documents →Extensible agent harness
Run Bkper CLI Agent in your terminal.
Bkper CLI Agent combines a lightweight agent harness with Bkper context, local tools, and included models. It ships with Bkper CLI through a terminal interface.
We use the Bkper CLI Agent every day to build Bkper itself.
Bkper AI
Choose the output your workflow needs.
Use a language model for conversations and generated results, or Jev when your code needs a boolean probability, one bounded choice, or an ordered score.
GPT · Luna, Sol 
Grok 
Gemini Flash 
GLM Flash 
DeepSeek Flash 
Jev Bkper CLI Agent uses the language models. Jev is an API model for apps and compatible clients, not a CLI Agent runtime. Learn about typed evaluations →
Ready to useSign in with your Bkper account
CuratedLanguage models we use to build Bkper
IncludedMonthly allowance, no automatic overages
AccountableUsage by user, model, and app
Start safely
Pick the smallest useful next step.
Start with one painful task. Keep writes reviewable and recurring work repeatable.
- Read Understand the rule
Learn why raw LLM output is a draft for finance work.
AI fundamentals → - Try Get one useful result
Ask a question, review a draft, or create a report before automating more.
Set up your assistant → - Build Turn wins into systems
Move recurring prompts into scripts, apps, reports, skills, or internal tools.
Build with agents →
Start free with reviewable AI workflows.
Connect Bkper to ChatGPT or Claude and turn prompts into reviewable work.