# Aurelio Finance: Self-Hosted Personal Finance with an AI Advisor

> Aurelio is a self-hosted personal finance application for tracking wealth, portfolios, debts, cash flow, and financial goals. Its optional AI chat uses OpenRouter and proposes changes for user confirmation.

**LosaLosSantos/aurelio-finance** — Open-source personal finance app with an AI financial advisor: track your net worth, investments, ETFs, cash and debts on your own computer.

- Repository: https://github.com/LosaLosSantos/aurelio-finance
- Stars: 349 · Forks: 104
- Language: Python
- License: MIT
- Published: 2026-10-16 · Updated: 2026-10-16 · Language: en
- Canonical page: https://hysenlabs.com/projects/losalossantos-aurelio-finance

## Aurelio's self-hosted wealth model

Aurelio addresses the problem of seeing household finances in one place. It is intended for people who track banks, brokers, cash, investments, property, loans, recurring contributions, and financial goals on their own computer. The README emphasizes local storage: records are kept in `backend/data.db`, and the application runs through a local startup script. That design differs from a cloud portfolio service because the user owns the deployment and the database rather than relying on a hosted account.

The application covers more than a quote list. Its wealth model includes dated situations for assets held at banks and brokers, while the portfolio section prices positions and records cost, gain, and dividends. Aurelio also includes plans, homes, and debts, plus cash-flow measurements such as income, expenses, savings rate, and the return required by each goal. This gives a personal finance user a connected view of liabilities and assets rather than treating investments as a separate activity.

Aurelio is not presented as a trading terminal. The README describes analysis, fund look-through, and an AI conversation, but it does not document order execution, broker authentication, or automatic transaction imports. Those omissions matter for evaluation. The project appears designed for maintaining a financial model and reviewing decisions, not for placing trades or replacing an official brokerage statement.

## Portfolio, fund look-through, and cash-flow calculations

The documented portfolio workflow begins with positions and their market prices, then adds the user's cost basis, gain, and dividend information. Aurelio also provides look-through views for the contents of funds by country, sector, company, and currency. That is useful when a user holds ETFs and wants to understand exposure that a simple ticker list does not reveal.

The README links to separate wiki pages for net worth, ETF overlap, portfolio X-ray, accumulating versus distributing ETFs, emergency funds, and AI advisor costs. This separation suggests that Aurelio is not just a dashboard generator. The documentation gives each calculation its own explanation, including how portfolio overlap is assessed and how accumulating and distributing ETFs differ. For a user reviewing a financial record, that documentation is more important than a claim that the interface is comprehensive.

Cash-flow analysis is tied to goals. The application records income, expenses, and savings rate, and can express the return each goal needs. This is useful for linking a current financial picture to a future requirement, such as funding a purchase or maintaining a planned investment allocation. However, the README does not specify the exact formulas, supported currencies, market-data refresh schedule, or behavior when prices are stale. Those details should be checked against the wiki and local data files before relying on a calculation for a consequential decision.

## How the AI advisor changes the workflow

The AI feature is optional. Aurelio uses OpenRouter with a key created by the user, and the README says that the key must be added to `backend/.env`. The AI chat can answer from the user's records, search the web, and propose a change as a card. The confirmation step is the important control. A proposed modification is not described as silently applied; the user is expected to confirm or reject it.

This creates a specific data flow. Financial records remain available to the application as its underlying data source. A question is sent through the configured AI service, potentially with a web search, and the result may include a proposed update. The interface presents that update for review before the application records it. The separation reduces the chance that a conversational instruction immediately mutates a portfolio, goal, or household record.

The model is not fixed by the local records. The README says the default is Claude Opus 5.5 and that another model can be selected in the chat. It also publishes example costs for questions, follow-ups, card confirmations, web searches, and an analysis. Those figures describe the README's October 2026 measurements on a test database, not a guaranteed production bill. OpenRouter usage, model choice, token volume, web searches, and the number of analysis actions will affect cost.

Because the AI is optional, Aurelio remains usable for local record keeping without an API key. That is a practical benefit for users who want to evaluate the financial model before sending any questions to an external service. The README does not describe every field sent to OpenRouter, so users who need strict data-control guarantees should inspect the backend configuration and request behavior before enabling the feature.

## Installing Aurelio with invented data

The repository provides a demo mode that starts with an invented household, which is the clearest first use for evaluating the application. The documented prerequisites are Git, Node.js 22 or later, and uv. On macOS or Linux, clone the repository and run the demo startup script. The first start installs its dependencies inside the folder and opens the local application at `http://localhost:8000`.

```bash
git clone https://github.com/LosaLosSantos/aurelio-finance.git
cd aurelio-finance
./start.sh --demo
```

The reader should see the local interface open with sample data rather than a personal financial record. The README does not document exact terminal output or installation duration, so those should not be treated as guarantees.

Windows users are directed to PowerShell. The documented command is:

```powershell
git clone https://github.com/LosaLosSantos/aurelio-finance.git
cd aurelio-finance
.\start.ps1 -Demo
```

If PowerShell blocks local scripts, the README gives this command:

```powershell
powershell -ExecutionPolicy Bypass -File .\start.ps1 -Demo
```

That bypass changes the script execution policy for this command invocation, so it should be used only when the repository script is trusted. After evaluating the demo, the user can start without the flag to work with their own data. The README says that personal records are stored in `backend/data.db`, and that Aurelio saves a copy next to it before an update changes the file format.

## Enabling OpenRouter and running local development

To enable the AI advisor, the README instructs the user to create an OpenRouter account, add credit, and create a key. The key is then placed in the local environment file. The documented configuration step copies `backend/.env.example` to `backend/.env`, after which `OPENROUTER_API_KEY` must be set.

The README does not provide a shell command that creates the key or edits the file, so the setup should be performed without adding undocumented commands. The application can still run without the AI feature, which makes it possible to separate local portfolio evaluation from OpenRouter account setup and billing.

Development commands are documented separately. Backend tests use `uv run pytest` from `backend/`, while frontend tests use `npm test` from `frontend/`. For hot reload, the backend uses `uv run uvicorn app.main:app --port 8000` in `backend/`, and the frontend uses `npm run dev` in `frontend/`. These commands indicate a Python backend and a Node.js frontend. The README does not state whether developers must start the frontend separately when using the normal startup scripts, so contributors should follow the repository instructions for their chosen workflow.

The MIT license permits use and modification under its conditions, but it does not replace responsibility for financial decisions, data backups, API billing, or compliance with third-party services. The README gives no warranty or data-recovery procedure beyond the stated database copy behavior.

## Limits of Aurelio versus a manual ledger

Aurelio's main trade-off is the amount of setup involved. A manual spreadsheet or paper ledger can represent assets and liabilities without a running application, but it does not provide Aurelio's connected portfolio, goal, debt, and AI views. Aurelio, in turn, requires Git, Node.js, uv, a local database, and, for AI use, an OpenRouter account with credit.

Aurelio also appears unsuitable as a replacement for a broker or bank portal when automatic reconciliation is required. The README does not mention connecting to financial institutions, importing statements, synchronizing transactions, or placing trades. That silence is a concrete limitation: users may need to enter or maintain data themselves. The README also does not describe a mobile client, multi-user access model, authentication system, or encrypted database format.

The local database is a clear operational boundary. It keeps records on the user's machine, but it also means backups, upgrades, file permissions, and recovery are the user's concern. The stated pre-format-change copy helps with upgrades, yet the README does not describe backup scheduling or what happens if `backend/data.db` is deleted.

The repository is not archived, and its last push was on 2026-10-16. That indicates recent repository activity, but the project has no GitHub releases, so upgrade cadence and release-level compatibility are not documented. Users should treat the `main` branch as a moving installation target rather than assuming a stable release process.

## Conclusion

Aurelio fits individuals who want a private, self-hosted view of household wealth and are willing to manage a Python and Node.js application. It is a poor fit for users who need a hosted service, a mobile-first interface, or automatic statement imports. Before adopting it, verify that your operating system can run the startup scripts, that its market and fund data sources match your holdings, and that OpenRouter costs fit your intended use of the AI features.

## FAQ

### What is Aurelio Finance used for?

Aurelio is a self-hosted personal finance and wealth-management app for tracking net worth, portfolios, ETFs, cash, debts, cash flow, and financial goals. It can also provide an optional AI advisor through OpenRouter.

### How do I run Aurelio with demo data?

The README documents cloning the repository and running `./start.sh --demo` on macOS or Linux, or `./start.ps1 -Demo` in PowerShell on Windows. The first start installs dependencies and opens `http://localhost:8000` with invented household data.

### Does Aurelio require OpenRouter?

No. The AI part is optional, and local records can be used without an OpenRouter key. To enable the chat, copy `backend/.env.example` to `backend/.env` and set `OPENROUTER_API_KEY`.

### Does Aurelio place trades or connect to my bank?

The README does not document trade execution, bank connections, broker authentication, or automatic statement imports. It describes a local application for tracking and analyzing financial records.

### Where are Aurelio's personal records stored?

The README states that personal records live in `backend/data.db`. It also says Aurelio saves a copy next to that file before an update changes the data format.

## Sources

- [Issues](https://github.com/LosaLosSantos/aurelio-finance/issues)
- [License: MIT](https://github.com/LosaLosSantos/aurelio-finance/blob/main/LICENSE)
- [LosaLosSantos/aurelio-finance on GitHub](https://github.com/LosaLosSantos/aurelio-finance)
- [README](https://github.com/LosaLosSantos/aurelio-finance/blob/main/README.md)

---

Hysen Labs editorial analysis, written from the project's own repository and release notes. Cite the canonical page: https://hysenlabs.com/projects/losalossantos-aurelio-finance