Slowly：記錄訂閲續期，避免意外扣款
Slowly: Remember Your Renewals Before Your Wallet Does
Slowly 是一款以本機為先的訂閲追蹤工具，可管理訂閲、查看預計每月開支及即將續期項目，並支援 JSON 備份和匯入。用戶可選擇以本機 AI 模型將自然語言描述整理成可編輯草稿；核心追蹤功能毋須 AI，資料儲存在瀏覽器中。
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady
What I Built
Everyday subscriptions create a small but persistent mental task: remembering when a free trial ends, when a plan renews, and when to cancel autopay before an unwanted charge. Slowly is a local-first subscription tracker that keeps those details in one calm, private place.
It lets people add, edit, archive, search, and filter subscriptions while showing estimated monthly spending, active subscription count, and upcoming renewals. It also supports JSON backup and import, so the data remains portable and under the user's control.
Slowly includes an optional natural-language entry workflow. A user can describe a subscription, receive a structured draft from a local AI model, review and edit the result, and then save it. AI is helpful, but it is never required for the core tracker.
Demo
hosted demo- https://harshdev-07.github.io/slowly--Hacktober/
The project can be also run locally with:
py -m http.server 4173
Then open http://localhost:4173.
Code
The complete source code is available here:
How I Built It
Slowly is built with vanilla HTML, CSS, and JavaScript ES modules. There is no npm dependency or build step. Subscription records are stored locally in the browser with IndexedDB, and a service worker plus web app manifest provide an offline-capable progressive web app experience.
The optional AI workflow connects to a local Ollama-compatible /api/generate endpoint. The default model is llama3.2:3b, but the endpoint and model can be changed in the app settings. The model turns a natural-language description into structured JSON. Slowly parses and validates that response locally, then presents it as an editable draft before anything is saved.
This design keeps the normal workflow usable without an account, analytics, cloud database, paid API, or exposed API key.
Why Does Open Innovation Matter?
Open innovation made it possible to build around local, open-weight AI instead of making a personal finance utility dependent on a closed hosted API. Users can choose the model that runs on their own machine, inspect how the response is handled, and keep subscription data in their browser.
The project is also intentionally easy to understand and contribute to. A contributor can open a small set of plain HTML, CSS, and JavaScript files without learning a framework build system first. That lowers the cost of improving a tool whose goal is to lower everyday cognitive load.
The Lazy Payoff
Slowly turns a recurring mental calculation into a quick dashboard check. Instead of keeping several renewal dates in memory or searching through emails when a free trial is ending, the user can see what is renewing soon, what each plan costs, and which payment details belong to it.
The payoff is small each day, but it compounds: fewer reminders to remember, fewer tabs to search, and fewer surprise renewal charges.
Prize Categories
- GitHub Copilot
來源：DEV Community · dev.to