jobbernaut-sensei 0.3.1

A spaced-repetition CLI for LeetCode practice problems

Jobbernaut Sensei

Navigate the vast space of the job market.

Jobbernaut is a cloud-native, event-driven open core intelligence platform designed to automate the most tedious parts of the job search: data extraction, resume tailoring, and document management.

Jobbernaut Sensei is the LeetCode practice & spaced-repetition CLI tool within the ecosystem — helping you sharpen your algorithmic skills with a structured review system. It's also an AI-agent-friendly API: every command outputs clean JSON, ready for agents like Cline, Claude Code, or ChatGPT to consume.

---

🚀 The Ecosystem at a Glance

Repository	Role	Technology Stack
jobbernaut-infra	Foundation	Terraform (HCL), AWS IAM, S3, DynamoDB
jobbernaut-tabs	Frontend	Next.js, React, Tailwind CSS
jobbernaut-backend	Control	Python (AWS Lambda), Boto3, Pydantic
jobbernaut-tailor	Intelligence	Docker, Python, LangChain (AI), TeXLive (LaTeX)
jobbernaut-extract	Collection	Chrome Extension (Manifest V3), JavaScript
jobbernaut-sensei	CLI Practice	Python (setuptools)

📘 Documentation: For deep dives into the architecture, data flows, and setup guides, visit jobbernaut-docs.

---

⚡️ Key Features

• Spaced Repetition: Never forget a solution — schedules reviews based on your difficulty rating.
• One-Command Scaffolding: Create new problem files with pre-filled metadata in seconds.
• ATS-Ready Tracking: Export your progress to CSV or Markdown.
• Context-Aware Fuzzy Matching: Open or mark problems by number, slug, or title words.
• Zsh Completions: Tab-complete problem names and categories.
• 🤖 AI-Agent Friendly: Every command outputs clean JSON — sensei revisit --json, sensei status, sensei show <problem>.
• Non-Interactive Marking: sensei mark 217 --rating e — perfect for agent-driven tutoring sessions.

---

📦 Setup

pip install jobbernaut-sensei

Initialize a problems/ directory:

sensei init

---

🔁 Daily Loop

1. Morning — see what's due

sensei revisit

2. Start a problem — scaffold it

sensei new 217 contains-duplicate 1-arrays-and-hashing -d easy -t arrays hash-set

3. Open a problem — jump back in

sensei open 217

4. After solving — mark it

sensei mark 217

You'll be prompted with one question:

  How did it go?

    [e]  easy      → 90 days
    [g]  good      → 30 days
    [h]  hard      → 7 days
    [s]  struggled → 3 days

Press one key. last_solved and revisit_in_days are updated automatically.

---

📖 Command Reference

sensei revisit — daily review runner

sensei revisit
sensei revisit --all
sensei revisit --topic arrays
sensei revisit --json            # AI-agent friendly JSON output
sensei revisit --export
sensei revisit --export-md

Colored terminal output:

📅  Jobbernaut Sensei Revisit — Tuesday, May 26 2026

🟢  UPCOMING (7 days)
──────────────────────────────────────────────────────────────────────────
  in 3d          [hard  ]   124. Binary Tree Maximum Path Sum      (trees)
  in 4d          [medium]   853. Car Fleet                         (stack, monotonic-stack)

──────────────────────────────────────────────────────────────────────────
  0 problem(s) need attention today.  Total tracked: 22

JSON output (--json):

{
  "generated": "2026-05-26",
  "total_tracked": 22,
  "problems": [
    {
      "label": "124. Binary Tree Maximum Path Sum",
      "difficulty": "hard",
      "last_solved": "2026-05-26",
      "due_date": "2026-05-29",
      "days_until_due": 3,
      "topics": ["trees"],
      "topic_folder": "7-trees",
      "filepath": "problems/7-trees/..."
    }
  ]
}

Flag	What it does
(none)	Overdue + due today + upcoming 7 days
--all	Everything, including far-future problems
--topic arrays	Filter by topic tag (partial match)
--json	Structured JSON output for AI agents
--export	Export all problems to export.csv
--export-md	Export all problems to export.md

---

sensei status — quick summary

sensei status

Returns a lightweight JSON snapshot of your tracker state:

{
  "total": 22,
  "overdue": 0,
  "due_today": 0,
  "upcoming": 4,
  "problems": [
    { "label": "124. Binary Tree Maximum Path Sum", "difficulty": "hard", ... }
  ]
}

Ideal for AI agents to quickly assess a user's practice state before a tutoring session.

---

sensei show <problem> — inspect a problem

sensei show 217
sensei show contains-duplicate

Returns metadata + your saved solution code as JSON:

{
  "label": "217. Contains Duplicate",
  "number": "217",
  "title": "Contains Duplicate",
  "filepath": "problems/1-arrays-and-hashing/...",
  "metadata": {
    "last_solved": "2026-05-14",
    "revisit_in_days": 90,
    "difficulty": "easy",
    "topic_tags": ["arrays", "hashing"],
    "due_date": "2026-08-12"
  },
  "solution": "class Solution:\n    def containsDuplicate(..."
}

Use this to fetch a problem for analysis, code review, or tutoring.

---

sensei new — scaffold a new problem

sensei new 217 contains-duplicate 1-arrays-and-hashing
sensei new 217 contains-duplicate 1-arrays-and-hashing -d easy -t arrays hash-set --open

Argument	Description
number	LeetCode problem number
slug	LeetCode URL slug, e.g. contains-duplicate
category	Topic folder, e.g. 1-arrays-and-hashing
-d / --difficulty	easy / medium / hard (default: medium)
-t / --tags	One or more topic tags
--open	Open the file in $EDITOR / code immediately

---

sensei open — open a problem

sensei open 217
sensei open contains-duplicate
sensei open 217 --no-browser   # editor only

Accepts problem number, slug, or title words — same fuzzy matching as mark.

---

sensei mark — mark a problem as reviewed

sensei mark 217
sensei mark contains-duplicate
sensei mark 217 --rating e     # non-interactive (AI-agent friendly)

Flag	What it does
(none)	Interactive prompt for difficulty rating
--rating e	Mark as easy (90 days until next review)
--rating g	Mark as good (30 days)
--rating h	Mark as hard (7 days)
--rating s	Mark as struggled (3 days)

The --rating flag skips the interactive prompt — perfect for AI agents tutoring a user and updating the schedule automatically.

---

🤖 AI-Agent Integration

See AGENTS.md — a complete tool-calling guide for integrating sensei into any AI coding agent's toolset.

Quick reference:

sensei status                     # Assess user's practice state
sensei revisit --json             # Get full review data
sensei show <problem>             # Inspect problem + solution
sensei mark <problem> --rating g  # Update schedule after tutoring

---

🔧 Zsh Shell Completions

Tab-complete sensei subcommands, problem numbers/names, and category names.

Add to ~/.zshrc:

fpath=(/path/to/jobbernaut-sensei/src/completions $fpath)
autoload -Uz compinit && compinit

Then reload: source ~/.zshrc

---

⚖️ License

Unless otherwise noted, all repositories within the Jobbernaut organization are licensed under the PolyForm Noncommercial License 1.0.0.

If a repository lacks a specific LICENSE file, the license located at Jobbernaut/LICENSE.md applies by default.

You are free to use, modify, and learn from this code for personal or non-commercial purposes.