Gemini Form Filler

Gemini Form Filler is a browser extension that uses the Google Gemini AI to automatically read and fill Google Forms on your behalf. Instead of spending time manually answering every question, you open the form, click Process, and let the extension handle the rest. HOW IT WORKS When you open a Google Form, the extension adds a Process button at the top of the page. Clicking it starts the process: the extension reads each question one by one, sends it to the Gemini API along with any available answer options, receives a generated answer, and fills it into the corresponding field automatically. Each question gets a status icon so you can see exactly what happened at a glance. The extension understands all major Google Forms question types: short answer, paragraph, multiple choice, checkboxes, dropdown, linear scale, rating, radio grid, checkbox grid, date, and time. File upload questions are automatically skipped since they require a real file from the user. QUESTION STATUS ICONS Every question gets a status icon that updates in real time as the extension works through the form: Pending — the question is queued and waiting to be processed. Processing — the question is currently being sent to Gemini and awaiting a response. Processed — an answer was successfully generated and filled in. Failed — the extension was unable to generate an answer after several attempts. Skipped — the question was intentionally skipped (see below). Already answered — the question already had an answer on the form itself before the extension ran. ALREADY-ANSWERED QUESTIONS If a question already has an answer when you click Process — filled in manually, or left over from earlier work on the form — the extension leaves it alone by default and marks it with a distinct icon, rather than silently overwriting it. A Reprocess button appears next to that question, so you can force the AI to answer it anyway whenever you actually want that. SMART SKIPPING Not every question has an objectively correct answer. Questions that ask for personal information — your name, email address, phone number, personal opinion, or individual experience — cannot be answered by an AI on your behalf. Instead of inventing a fake answer, the extension detects these questions and skips them automatically, marking them with the Skipped icon. This way you always know which fields still need your attention before submitting. File upload questions are also skipped automatically. IMAGE SUPPORT Some Google Forms questions include images — for example, a math problem written on a diagram, a chart you need to interpret, or a picture you need to describe. Gemini Form Filler supports these questions fully. When a question contains an image, the extension fetches it and includes it in the request to Gemini alongside the question text. Gemini analyzes both the text and the image together to produce the answer. RATE LIMIT HANDLING The Gemini API has rate limits on how many requests can be made per minute. If the extension hits a rate limit mid-form, it does not fail silently — it pauses, shows a banner with a countdown timer indicating how long to wait, and automatically resumes from the exact question where it stopped once the limit is lifted. No progress is lost, and no questions need to be re-processed manually. If your API quota is fully exhausted for the day, the extension notifies you immediately, marks all remaining questions as failed, and stops — so you are never left wondering why answers stopped appearing. RE-RUNNING PROCESS If some questions failed on the first attempt, you do not need to re-process the entire form. Clicking Process again will only attempt questions that are currently marked as failed or pending. Questions already marked as processed or skipped are left untouched, and already-answered questions stay under your control via their own Reprocess button, as described above. INTERFACE LANGUAGE The extension's own interface — buttons, status messages, tooltips — is available in multiple languages and automatically follows your browser's language setting. Currently supported: English and Ukrainian. The project is open source, and anyone is welcome to contribute a translation into their own language; see the README on GitHub for instructions. This is separate from the language Gemini answers in, described next. SETUP To use the extension, you need a Google Gemini API key. You can get one for free at Google AI Studio (aistudio.google.com). Once you have a key: 1. Click the extension icon in your browser toolbar to open the settings popup. 2. Paste your API key into the input field. 3. Click Load models. 4. If the key is valid, a dropdown will appear listing all Gemini models available for your account. 5. Select the model you want to use. 6. Your settings are saved automatically and persist between sessions. The model list refreshes every time you open the popup. If a model you previously selected is no longer available under your account, the extension will notify you and ask you to choose a new one. Once everything is set up correctly, the popup stays out of your way — the Load models button only reappears if something actually needs your attention again. CHOOSING A MODEL The dropdown only lists stable, general-purpose Gemini models capable of generating text — experimental, preview, and specialized (image-only, embedding-only, etc.) variants are filtered out automatically, so you are not left guessing which option is safe to pick. Models are sorted with Flash variants at the top since they are the fastest and most cost-effective for this use case. For most forms, a Flash model is a good default choice. ANSWER LANGUAGE The extension instructs Gemini to always respond in the same language the question is written in. If a question is in Ukrainian, the answer will be in Ukrainian. If it is in English, the answer will be in English. No manual language configuration is needed, and this works independentl…