Privacy Policy
Trader Tools is a personal financial dashboard that I built and operate. It brings together budgeting, debt tracking, tax records, and investment portfolio data in one place. This policy explains exactly what information the application collects, where that information goes, and how you can remove it.
1. Information collected
Account and identity
Sign-in is handled entirely by Google. The application never sees, receives, or stores a password. From Google's sign-in response it stores your Google account identifier, email address, display name, and profile picture URL.
Financial information you enter
Anything you record in the application is stored in its database, including: asset accounts and balances; debts, with interest rates, balances, and payment amounts; income sources, pay rates, and pay schedules; recurring bills; budget categories and limits; individual spending transactions, including merchant, amount, date, and category; investment holdings, trades, cost basis, and dividends; and any notes you attach to these records.
Documents you upload
Tax documents and receipts you upload are stored as complete files in the database. This includes documents such as W-2 and 1099 forms, which typically contain Social Security numbers, employer details, and full wage information. Uploaded files are retained until you delete them.
Connected financial accounts
If you choose to connect a bank or brokerage account, that connection is made through Plaid. Your banking credentials are entered directly into Plaid's interface — they are never transmitted to, seen by, or stored by this application. What is received back and stored is account metadata, balances, transaction history, and, depending on the account type, loan terms and investment holdings.
Technical information
A session cookie is set to keep you signed in. Server logs record request paths, timestamps, and error details for operational and diagnostic purposes. If you create a personal access token for API use, a record of that token is stored.
2. How information is used
Information is used solely to operate the features you are using: displaying your financial position, calculating budgets and cash-flow projections, tracking portfolio performance, estimating taxes, and generating written analysis at your request. It is not used for advertising, profiling, or any purpose unrelated to running the application, and it is not sold or licensed to anyone.
3. AI processing
Several features send your data to third-party AI providers for processing. This is the most significant way information leaves the application, so it is described in detail here rather than buried in a list.
- Written financial analysis. When you request an AI read — a financial outlook, a tax summary, a portfolio or budget analysis — a structured summary of the relevant figures is transmitted to Anthropic (Claude) and, for some features, to Google (Gemini). This summary contains real financial values such as balances, debts, income, and holdings.
- Document reading. When you upload a tax document or receipt, the complete file is transmitted to Anthropic so its contents can be extracted automatically. For a W-2 or 1099, that means the full document — including any Social Security number printed on it — is sent to that provider. If you would rather this not happen, do not use the automatic extraction feature; the corresponding fields can be entered by hand instead.
- Results are cached. AI-generated text is stored in the database and reused for a period of time so identical requests do not trigger a new call.
- Local processing option. The application can fall back to a locally-hosted language model. When that path is used, the data involved does not leave the server.
4. Where data is stored
Data is stored in a PostgreSQL database running on infrastructure I operate. The application is served over HTTPS with certificates issued and renewed automatically. Application credentials are held as Kubernetes secrets and are not present in source code or container images. The application container runs as an unprivileged user. Database backups are taken on an automated daily schedule and retained for 30 days.
5. Service providers
These are the third parties that receive data, and what each one receives. None of them receive your information for their own independent purposes.
| Provider | What it receives | Why |
|---|---|---|
| Sign-in identity: account ID, email, name, profile picture | Authentication | |
| Anthropic (Claude) | Financial summaries; complete uploaded tax documents and receipts | Written analysis and document field extraction |
| Google AI (Gemini) | Financial summaries | Secondary analysis on some features |
| Plaid | Your banking credentials, entered directly into Plaid — never through this application | Connecting financial accounts, where you choose to |
| Market data providers | Ticker symbols only — no personal or account information | Prices, fundamentals, and news |
Plaid handles your information under its own privacy policy, available at plaid.com/legal.
6. Retention and deletion
Your records are kept until you delete them or delete your account. Deleting your account removes personal details from the active record immediately and disables sign-in; the account can be restored by signing in again within the retention window, after which it can be permanently purged. Individual records — transactions, documents, accounts, debts — can be deleted at any time from within the application, and deletion is immediate.
One limitation worth stating plainly: deleted data may persist in database backups for up to 30 days after deletion, until the backup containing it ages out of the retention window.
7. Cookies
A single session cookie keeps you signed in. There are no advertising cookies, no analytics trackers, and no third-party tracking of any kind.
8. Your choices
- View and edit any record you have entered, directly in the application.
- Delete individual records, uploaded documents, or your entire account.
- Export your portfolio and tax data.
- Disconnect a linked financial institution, which stops further data retrieval.
- Decline AI document extraction and enter document fields manually instead.
9. Children
This application is not directed to children and is not intended for use by anyone under 18. No information is knowingly collected from children.
10. Changes to this policy
If this policy changes, the revised version will be posted here with an updated date at the top. Material changes to how data is shared will be reflected here before the change takes effect.
11. Contact
Questions about this policy, or requests regarding your data, can be sent to admin@kegbot.net.