SecondHand

SecondHand stores your details on your own computer and autofills support forms in Chrome, starting with Iowa SNAP, so people spend less time retyping and more time getting help.

Comment

Inspiration

SecondHand was inspired by a simple but frustrating reality: people seeking food and other assistance often have to re-enter the same personal information across multiple forms and programs. The work is not only repetitive, it is emotionally draining. When someone is already stressed, time is one of the most valuable resources they do not have.

This project was built for Hack Away Hunger 2026 with the goal of reducing that burden while respecting privacy and keeping the person in control. We wanted to create something that helps people fill out forms once, instead of repeating the same answers over and over, without forcing them to hand their data to a cloud service or trust a system that decides eligibility for them.

The project also grew from the idea that people deserve a better experience when they are trying to access food, housing, and other support. The barrier is often not a lack of help, but the administrative friction of the application process itself.

What it does

SecondHand is a privacy-first desktop app and Chrome extension that helps users fill out benefit applications with less manual work. The app stores a person’s details locally on their own computer, and the extension can autofill supported fields on websites such as Iowa SNAP and other assistance forms.

The workflow is intentionally simple:

  • A user saves their profile once
  • The app stores the information locally and securely
  • The extension detects supported forms in Chrome
  • The user clicks Autofill
  • The app fills in known answers and leaves the final review to the user

It is designed to reduce repeated typing, not replace human judgment. It does not decide if someone qualifies. It does not submit applications for them. Instead, it helps complete the repetitive parts while keeping the user in charge.

How we built it

SecondHand was built as a combination of a desktop application and a browser extension. The desktop app stores encrypted profile data on the user’s own machine, while the browser extension communicates with it locally to fill fields in supported websites.

The architecture includes:

  • Electron for the desktop app
  • Chrome extension APIs for browser integration
  • Encrypted local storage for personal information
  • A generic form-matching adapter for supported websites
  • A hand-mapped flow for high-value applications such as Iowa SNAP
  • Optional local AI for ambiguous questions when confidence is high
  • OCR and document parsing for PDFs and photos of tax forms

The project also emphasizes on-device privacy. There is no account requirement, no cloud copy of the user’s personal data, and no backend dependency for the main workflow. That was a deliberate design choice to keep the product trustworthy and useful for vulnerable users.

Challenges we ran into

One of the biggest challenges was balancing automation with safety. Autofill can save a lot of time, but it can also create risk if the wrong answer is inserted or if sensitive data is exposed in the wrong place. To address this, we designed the app to require user approval, only fill recognized fields, and keep final review in the user’s hands.

Another challenge was form variability. Different websites ask the same questions in different ways, with different labels, layouts, and sequencing. We had to build logic that is both general enough to work across websites and specific enough to handle high-value forms reliably.

We also had to think carefully about trust and ethics. If a user is already stressed or overwhelmed, they need the tool to feel helpful, not invasive. Privacy, clear prompts, and transparent behavior became central to the product, not just add-ons.

Accomplishments that we're proud of

We are proud to have built a working prototype that demonstrates the core idea: people can save their information once and reuse it across form-heavy assistance processes.

Some accomplishments we are especially proud of include:

  • Building a working autofill flow for Iowa SNAP applications
  • Keeping personal data on-device rather than in the cloud
  • Creating a browser extension that works with existing websites without requiring agency changes
  • Supporting multiple languages and simplifying complex pages for users
  • Adding local document parsing for PDFs and photos
  • Designing the system to be privacy-first, accessible, and human-centered

These are meaningful wins because they show that the project is not only technically feasible, but also aligned with the real needs of the people it is meant to serve.

What we learned

The project reinforced that the hardest part is not building a form autofiller, but building trust. People need to feel safe handing sensitive information to a tool, especially when they are already under pressure. Privacy has to be a core feature, not a later improvement.

We also learned that the most valuable automation is the kind that removes tedious work without taking away agency. SecondHand is most effective when it handles the repetitive parts and leaves judgment, review, and decision-making to the user.

Another key lesson was that real benefit forms are messy. They vary in structure, ambiguity, and edge cases. A successful solution needs a mix of strategy: deterministic rules for highly structured flows, broad generalization for other sites, and carefully limited use of AI for ambiguous cases.

What's next for SecondHand

The next step is to expand support beyond the current high-value workflows and improve the product’s general usefulness across more assistance forms. We want to continue refining the form-matching logic, expand the number of supported websites, and improve the user experience for both first-time and repeated users.

Longer-term goals include:

  • Broader support for more food and income-assistance forms
  • More reliable handling of unusual or ambiguous question wording
  • Improved user onboarding and recovery flows
  • Better support for shared or public computers
  • Continued focus on privacy, accessibility, and low-barrier adoption

The mission remains the same: reduce administrative friction, preserve privacy, and make it easier for people to access the support they need without burning time on the same questions over and over.

Built With

Share this project:

Updates

Submission history