GitHub - jiito/interview-prep-skills: a few agent skills I have made to help study for interviews

Cursor-compatible Agent Skills for coding and system design interview practice. Each skill extends your AI assistant with interview-specific workflows: drills, prompt generation, interactive practice, and design review.

flowchart TB
    subgraph skills [Interview Prep Skills]
        RPD[requirements-prioritization-drill]
        SDI[system-design-interview]
        IG[interview-generation]
    end

    RPD -->|"5-10 min reps"| Scope[Requirement scoping]
    SDI -->|"Generate / Practice / Review"| Design[System design]
    IG -->|"4-part Python prompts"| Coding[Coding interviews]

Skills

requirements-prioritization-drill

Rapid reps-based drill for the first 5–10 minutes of a system design interview: parsing ambiguous prompts, asking follow-up questions, and compressing the problem into prioritized functional and non-functional requirements.

Use when: You want to practice follow-up questions, requirement scoping, or turning an ambiguous prompt into a small set of prioritized requirements.

Example prompts:

• "Practice system design follow-up questions"
• "Run a quick requirements drill"
• "Help me get better at requirement prioritization"

Capabilities: Ephemeral, chat-only rounds with immediate feedback on what was strong, what to tighten, and ideal follow-up questions. No files created unless you ask.

system-design-interview

Generate system design practice (prompt + solution files), run interactive interview sessions, and review whiteboard designs via Excalidraw.

Use when: You want to create, practice, or get feedback on a system design interview.

Example prompts:

• "Generate a new system design interview"
• "Let's practice the [topic] system design question"
• "Review my design for [problem]"

Capabilities:

• Generate: Creates sd_{name}_interview.md and sd_{name}_solution.md with realistic prompts, time allocation, and reference answers. Researches hellointerview.com and avoids duplicate topics.
• Interview: Acts as interviewer, answers clarification questions in character, never dumps the solution.
• Review: Reads your filled sections and Excalidraw diagram; compares against reference; gives rubric scores and actionable feedback.

interview-generation

Generates Python coding interview practice prompts in a 4-part format: Part 1 (core implementation), Part 2 (extension), Part 3 (deeper twist), Part 4 (discussion). Produces runnable skeletons with type hints and optional tests.

Use when: You explicitly ask to create or refine interview practice questions/prompts.

Example prompts:

• "Generate a new Python coding interview question"
• "Create an implementation-focused interview prompt"
• "Add a new algorithm-leaning interview with tests"

Capabilities: Scans repo for existing questions to avoid duplicates. Supports implementation-focused, algorithm-leaning, or mixed styles. Optional test generation with fake clock injection. Real-world framing, explicit requirements, and evaluation criteria table.

Installation

With npx skills (recommended)

You can install all skills at once using npx skills:

npx skills add jiito/interview-prep-skills

Skills Format

Each skill follows the Agent Skills specification with a SKILL.md file containing frontmatter (name, description) and instructions. Compatible with Cursor, Claude Code, and other platforms that support the open standard.