About Agents Finance
A directory for people who work in finance or run a small business and want to know what other people and companies really do with AI and agents in accounting, treasury, consolidation, annual accounts, controlling and reporting, and in the ERPs they use.
Who it is for
Finance professionals, controllers, finance and administration managers, accounting and advisory firms, and small and mid-sized businesses in English- and Spanish-speaking markets. The tone is practical: who does it, with which tools, what result is reported, what has really been tested, and how you could replicate it.
What we publish
The stars are the real use cases. Tutorials, guides, prompts and community resources are there to help you replicate them. Product news and vendor announcements are context, and we keep them to a minority.
- Use cases: An identifiable person, firm or company explains what they do with AI or agents in finance: with which tools, how, and with what result.
- Tutorials: Steps to do something specific with AI in finance or with an ERP, written by someone who has done it.
- Guides and manuals: Longer documents that explain how to set up, configure or control a use of AI in finance.
- Useful prompts: Ready-to-copy instructions you can adapt, with their variables and usage warnings.
- Community resources: Open tools, templates and repositories the community uses with AI in finance and ERPs.
- News: A relevant launch, announcement or study. It is context: we prioritize what people already use.
What a use case includes
Who does it (an individual, firm, company or organization), the sector if the source says so, tools and ERP, finance process, the reported result, how to replicate it, the level of evidence and a section called “What has really been tested”. If the source does not give a piece of data, we do not add it.
Prompts and confidential data
Prompts are published in a copyable block, with their variables and a warning: do not paste confidential or personal data into an AI service without checking your provider’s terms and data-protection rules (GDPR in the EU), and always review the output. A prompt we wrote ourselves is labeled “untested proposal” until someone tries it with real data.
Levels of evidence
Every entry states how much confidence it deserves, so an announcement is not mistaken for a result, nor a personal experience for proof:
- Real case: An identified person or organization states that they have used it. The results are stated by the source; we have not audited them.
- First-hand experience: Someone 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 proposal: Material written by this directory based on published practices. We have not tested it with real data: try it with sample data first.
- Project documentation: Described 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.
- Announcement: A communication from the vendor itself or from a news outlet about a launch. It describes what the product promises, not verified results.
- Opinion / analysis: A study, survey or analysis by a third party. Useful as context; it does not prove that a specific tool works.
How we write the entries
- Every entry starts from a public source with a verifiable link and its real publication date. We prefer testimonials from people and companies, repositories and practical manuals over press releases.
- We write in our own words and do not include figures or quotes the source does not confirm.
- If a source does not give the exact day or does not report measured results, we say so.
- The directory is produced with the help of AI tools for research and drafting, and every entry shows the date on which its source was last checked.
- Corrections are published as soon as they are confirmed. You can send us one.
English and Spanish
The site is published in English (the main language) and in Spanish. Each entry exists in both languages when a translation is available; the version in the other language keeps the same sources, figures, warnings and level of evidence, and translations are prepared with the help of AI tools. If an entry exists in only one language, we do not offer a fake version in the other. The language switcher is in the header.
Independence and sponsorship
Content is chosen for its usefulness to the reader. If there is ever sponsorship, it will be visible, labeled, and will not change the level of evidence of any entry. See the sponsors page.
For AI assistants
We publish /llms.txt (English) and /es/llms.txt (Spanish), with full-text versions at /llms-full.txt and /es/llms-full.txt, plus an RSS feed per language. AI crawlers are welcome to access the site; we ask that you cite the original source of each entry and this directory.
Last review of this page: October 17, 2026.