jonathanpopham/syllabus

Syllabus is a tool that turns a job posting into a personalized interview prep course. You give it a link to a job listing, and it produces a study package built specifically for that role: a primer on the company and its product and culture, plus a module for each stage of the interview process. Each module includes study notes, a quiz, and coding exercises written in whatever programming languages the job posting actually mentions. Instead of being a standalone app, Syllabus is built as a skill, meaning it runs inside an AI agent such as Claude Code, Claude.ai, or Hermes. The repository itself does not call any AI service directly. It provides instructions, a script for rendering the final course into an HTML file, and reference material the agent uses while building the course. The project is not affiliated with Udemy, the word course refers only to the format of modules and lessons, not any platform. When you use it, the agent researches the target company beyond the job description, looking at its product, business model, and any published information about its actual interview process. It then chooses a stage by stage structure to match, using the company's real process where it can find one, or falling back to a default four stage ladder. For each stage, it writes real, runnable exercise files, including a starter file, a test, and a solution, scaled in difficulty and grounded in the company's actual domain rather than generic algorithm questions. Everything is assembled into a course.json file and a self contained course.html file that opens in any browser with no external dependencies, so it is safe to view offline or share with a study partner. Installation is done by cloning the repository and running an install script that links the skill into whichever agent you have installed, either Claude Code or Hermes, or both. Setup takes only a few minutes, and the license is MIT.