Skip to content
Agents Finance

5 real use cases · 18 entries

What people and companies actually do with AI and agents in finance and ERPs

Real use cases in accounting, treasury, consolidation, annual accounts, controlling and reporting: who does it, with which tools and with what reported result. With the tutorials, guides, prompts and resources to replicate them.

Try:ReconciliationPeriod-end closeInvoicesMCP and connectors

Use cases
5
ERPs covered
7
Last reviewed

Real use cases

Identifiable people, firms and companies who explain what they did and with what result.

See all 5 use cases
  • Use caseEvidence: Real case

    Ten weeks keeping the books of Swedish limited companies with Claude Code and Codex, with wins and mistakes

    Philip Eriksson (co-founder of Igdrasil)Individual · Swedish limited companies in software, e-commerce, digital health, services and food

    Philip Eriksson, co-founder of Igdrasil, explains how, from mid-July 2026, he kept the books of several Swedish limited companies by talking to Claude Code and Codex. He provides volumes taken from his logs, a VAT error caught before filing and a table of agent failures. He is the author of an accounting tool.

  • Four-step diagram of UNHCR's cash forecast: accounts payable, future orders and history data; date normalisation; model and dashboard; and human review and decision in treasury.
    Use caseEvidence: Real case

    UNHCR uses AI to prepare its treasury cash forecast: what it did, what it gained and how to replicate it

    UNHCR, the UN Refugee Agency — treasury unitOrganization or institution · International humanitarian organisation (United Nations system)

    The treasury team at UNHCR, the UN Refugee Agency, explained at EuroFinance 2026 how it uses AI to collect, normalise and project its cash flows, with treasurers still responsible for reviewing and deciding. It is a real case told by its head; the 40% to 60% reduction in work is his estimate for once the model is stable, not an audited result.

    • AI cash forecasting model developed with UNICC (the source does not detail the type of model)
    • ERP data: accounts payable and future-dated purchase orders
    • Almost ten years of cash flow history
    • +2
    • Treasury
    • Reporting

  • Four-step diagram: a Slack command reaches the AI agent, which reads the ERP general ledger and prepares a draft workpaper and journal entry that a human preparer and reviewer post.
    Use caseEvidence: Real case

    A CFO explains how their team reconciles prepaid expenses with an AI agent: from 2 hours to 5 minutes

    Author of OnlyCFO (a company CFO, no name visible on the page)Individual

    The author of the OnlyCFO newsletter, who runs an accounting team, describes the agent they built to reconcile prepaid expenses against their ERP’s general ledger. According to them, it went from about 2 hours of human work to about 5 minutes, and the same pattern took a day off the monthly close.

    • AI agent with a “skill” (a SKILL.md of about 200 lines) and a YAML configuration file
    • MCP connector to the ERP (the article does not say which one)
    • MCP connector to the shared drive (Google Drive, Box or SharePoint)
    • +2
    • Accounting

  • Use caseEvidence: Real case

    A tax and accounting advisory firm in Madrid automates invoice intake with n8n and ChatGPT

    FISCONSULTING, tax and accounting advisory firm in Madrid (project by Álvaro Díaz Blanco, UPM)Accounting or advisory firm · Tax and accounting advisory

    A final-year thesis at UPM describes two automations built in n8n for FISCONSULTING, an advisory firm in Madrid: one that extracts and classifies invoices from Google Drive with ChatGPT, and another that handles appointments over WhatsApp. The author states that the time per invoice drops from 2 minutes to about 30 seconds.

    • n8n (on a Hostinger cloud server)
    • Google Drive and Google Sheets
    • PDF.co
    • +3
    • Accounting

  • Use caseEvidence: Real case

    U.S. AutoForce saves more than 30 hours a month on reconciliations with Copilot for Finance

    U.S. AutoForce (U.S. Venture)Company

    U.S. AutoForce, a division of U.S. Venture that uses Dynamics 365, automated its recurring credit card and bank statement reconciliations with Microsoft Copilot for Finance. According to Microsoft, it took four weeks to roll out and its accounting team saves more than 30 hours a month.

Put it into practice

Tutorials, guides, prompts and resources people use today to replicate those cases.

Explore by type

  • Tutorials

    Steps to do something specific with AI in finance or with an ERP, written by someone who has done it.

    1 entry

  • Guides and manuals

    Longer documents that explain how to set up, configure or control a use of AI in finance.

    1 entry

  • Useful prompts

    Ready-to-copy instructions you can adapt, with their variables and usage warnings.

    1 entry

  • Community resources

    Open tools, templates and repositories the community uses with AI in finance and ERPs.

    1 entry

  • Community resourceEvidence: Project documentation

    Open-source MCP server to connect AI assistants to Odoo: read, aggregations and, if enabled, write

    ivnvxd and project contributors on GitHubIndividual

    A community project on GitHub (MPL-2.0 license, more than 400 stars, latest release published on 10 October 2026) that lets Claude and other assistants query and modify data in an Odoo instance: invoices, customers, products and aggregations. It has a test mode that needs no module and a module for production.

    • MCP server in Python (launched with uvx)
    • Odoo 16 or later (MCP module for production; test mode over XML-RPC)
    • Claude Desktop, Claude Code, Cursor or another MCP client
    • Odoo
    • Accounting
    • Reporting

  • Useful promptEvidence: Untested proposal

    Prompt for an AI assistant to prepare a draft accounting reconciliation without posting anything

    Instructions ready to adapt for an AI assistant that reconciles an account against the trial balance: it uses only the files provided, posts no entries, checks for duplicates, separates what ties out from what is doubtful and leaves a list of questions. It is our own wording based on practices published by accountants, not tested with real data.

    • Any AI assistant with file access (for example, Claude, ChatGPT or Codex)
    • Exports from your ERP or bank (CSV or Excel)
    • Accounting
    • Treasury

  • Guide / manualEvidence: First-hand experience

    Guide for bookkeepers: Claude as an assistant configured with a CLAUDE.md, what to delegate and what to always review

    Michael Robichaud (Cowork Consultant)Individual · Bookkeeping and small firms (with US examples, such as 1099 forms)

    Michael Robichaud, a consultant who rolls out Claude at accounting firms, explains how to set up the rules file (CLAUDE.md), which tasks the assistant can prepare (reconciliation, categorization, client reminders, close) and which always need human review. It is a consultant’s guide with an associated commercial service.

    • Claude (workspace with memory, files and a CLAUDE.md file)
    • Bank and ledger exports
    • Aging reports (AR/AP)
    • Accounting
    • Treasury

Full library: search and filter

18 entries

By ERP

See all ERPs

By finance process

See all processes

How to read the level of evidence

An announced feature is not an achieved result, and one person’s experience is not general proof. That is why every entry carries a label and a section on what has really been tested.

Read our editorial criteria

  • Real caseAn identified person or organization states that they have used it. The results are stated by the source; we have not audited them.
  • First-hand experienceSomeone explains how they do it or did it, without verifiable figures. It is documented practice, not proof that it will work in your case.
  • Untested proposalMaterial written by this directory based on published practices. We have not tested it with real data: try it with sample data first.
  • Project documentationDescribed by the project’s own documentation or README. It says what the project claims to do and how it is used, not that it works in your case: we have not tested it.
  • AnnouncementA communication from the vendor itself or from a news outlet about a launch. It describes what the product promises, not verified results.
  • Opinion / analysisA study, survey or analysis by a third party. Useful as context; it does not prove that a specific tool works.

Do you know a real use case that should be here?

Send us the link to the public source and we will review it. If it is yours, we show it as such.

Suggest an entry