ApplyPack

ApplyPack is an open-source job search tool by Nazar Boyko. Once installed, you use it in the browser on your own computer at 127.0.0.1:4747. It pulls listings on a schedule from the official public APIs of hiring systems such as Greenhouse and Lever, and from RSS feeds. The README says the AI reads the full description against your stack, role types, seniority, countries and salary floor. For a role you like, it can prepare an application pack. A fact check blocks numbers and experience that have no source, and AI cover letters use only facts you have confirmed. The README states that nothing is ever submitted for you; you apply yourself. It can tell you about new roles through Telegram or Discord.

Who it's for:Job seekers who can install Node.js and want a tool to keep watch on public job listings. Installation uses the command line, which will be hard going if you have not done that before.

Before you start: You need Node 22 or later on macOS, Windows or Linux. The documentation is in English, and the interface languages are not stated. Running npm install triggers a Prisma generate step, and some dependencies may download native binaries for your system. npm start builds the app, then starts a local PostgreSQL database plus the web and worker processes. Your data lives in a local PostgreSQL database in your user data folder, or in a PostgreSQL server you host yourself. After setup, once you switch it on, it checks job boards every hour, with an optional daily digest and quiet hours. The generic company career page source reads public web pages to find feeds or watch for text changes; it does not use browser login cookies or get around captchas. The AI can be a model provider, a command-line tool or a local model of your choice, and cloud models receive the relevant data. Automatic preparation is off by default, and when on it only prepares files and submits nothing. AI Inspiration World recommends only the job seeker features and does not recommend Employer mode, which screens other people's resumes automatically. The software is free, with no subscription or hosted account; you pay for any AI subscription or tokens. MIT license.

Based on the project's public documentation. Not tested by AI Inspiration World.