LeetCode Sensei

An OverKill Labs spaced-repetition CLI for LeetCode practice.

---

Your LeetCode practice, but you actually remember what you solved.

Sensei is a spaced-repetition CLI that stops you from grinding LeetCode into the void. Every command outputs clean JSON — it's built for humans but designed for AI agents.

pip install leetcode-sensei
sensei init

---

🔥 Why This Exists

You solve 200 LeetCode problems. Three months later you can't solve FizzBuzz.

Sensei tracks when you solved a problem and how well you understood it. Then it tells you exactly what to review — before your brain evicts it from cache.

┌──────────────────────────────────────────────────────────────┐
│  Your brain has a cache. This is the invalidation policy.   │
└──────────────────────────────────────────────────────────────┘

Zero config. Zero cloud. Zero bullshit. Your solutions are just .py files in a folder — git-friendly, portable, no lock-in.

---

⚡ What It Does

Command	Vibe
sensei revisit	"Here's what you're forgetting" — color-coded review queue
sensei status	JSON snapshot for you or your AI agent
sensei hint <problem>	Get the problem URL + deets without the answer (for quizzing)
sensei show <problem>	Full problem + your saved solution (for review)
sensei mark <problem> --rating g	"Good" → next review in 7 days (with load smoothing)
sensei new <num> <slug> <cat>	Scaffold a fresh problem in 0.3 seconds
sensei open <problem>	Jump into the code + LeetCode page
sensei rebalance	Spread review clusters across empty days
sensei revisit --export	Dump everything to CSV or Markdown

---

📅 Daily Loop

# 1. What's rotting in my brain today?
sensei revisit

# 2. Quiz me on this one (agent, don't spoil it)
sensei hint 217

# 3. Let me code, then compare
sensei show 217

# 4. How'd I do?
sensei mark 217 --rating g    # good → 7 days

That's it. Four commands. Whole loop takes 2 seconds of typing.

---

🧠 Spaced Repetition — The Smart Part

Rate yourself after each solve. Intervals are hardcoded — no adaptive math, no history weighting:

Rating	Flag	Next Review
Trivial 🏆	--rating t	90 days
Easy 🟢	--rating e	30 days
Good 🔵	--rating g	7 days
Hard 🟡	--rating h	3 days
Struggled 🔴	--rating s	1 day

The schedule is the curriculum. Every rating directly sets the next review — no surprises.

📈 New-Problem Progression Gate

New problems are gated through a 1 → 3 → 7 → 30 day ladder before entering full SRS. No matter how well you solve a brand-new problem, it can't skip ahead:

Review #	Max interval
1st solve	1 day
2nd solve	3 days
3rd solve	7 days
4th solve	30 days
5th+	Full SRS (no cap)

Rating e on your first solve → capped at 1 day. Rating s on your 4th solve → 1 day (under cap). The gate only fires when the rating would exceed the cap.

⚖️ Load Smoothing

sensei mark automatically spreads reviews across nearby days to avoid spikes. Instead of 6 problems all landing on the same day, the scheduler finds the least-loaded day within a window:

Rating	Base	Spread range
s	1d	fixed
h	3d	2–4 days
g	7d	5–14 days
e	30d	15–45 days
t	90d	45–90 days

Problems reviewed 5+ times are biased toward the later end of their window (stable memory, safe to defer). Newer problems stay at the earlier end (fragile memory, keep close).

Disable with --no-spread to get exact base intervals.

---

🤖 AI-Agent Native

Every command returns clean JSON. Plug it into Cline, Claude Code, ChatGPT, or your own agent:

sensei status                     # → {"total":22,"overdue":0,...}
sensei revisit --json             # → Full review data with dates
sensei hint 217                   # → Problem URL + status, no solution
sensei show 217                   # → Metadata + saved solution
sensei mark 217 --rating e        # → Update schedule, no prompts

See AGENTS.md for the complete agent integration guide.

---

📦 Setup

pip install leetcode-sensei
sensei init                      # Creates problems/ directory
sensei new 217 contains-duplicate 1-arrays-and-hashing -d easy -t arrays hash-set
code problems/                   # Start solving

Dependencies

• Python ≥ 3.10
• That's it. No npm. No Docker. No cloud.

---

📖 Commands

sensei revisit — what to review

sensei revisit                    # Overdue + due today + upcoming 7 days
sensei revisit --all              # Everything
sensei revisit --topic trees      # Filter by topic
sensei revisit --json             # Agent-friendly JSON
sensei revisit --export           # → export.csv
sensei revisit --export-md        # → export.md

Colored terminal output:

📅  LeetCode 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)

sensei show <problem> — inspect + solution

sensei show 217
sensei show contains-duplicate

Returns JSON: label, number, title, filepath, metadata (last_solved, revisit_in_days, difficulty, topic_tags, due_date), and your saved solution code.

sensei hint <problem> — inspect, NO solution

sensei hint 217
sensei hint contains-duplicate

Same as show but no solution field. Returns times_reviewed so the agent knows review history without seeing the answer. Perfect for agents that want to quiz without spoiling.

sensei mark <problem> — rate your session

sensei mark 217                         # Interactive prompt
sensei mark 217 --rating e              # Non-interactive (agent-friendly)
sensei mark 217 --rating g --no-spread  # Exact base interval, no smoothing

Automatically increments times_reviewed and applies the progression gate and load smoothing.

sensei rebalance — flatten review spikes

sensei rebalance              # Dry run — preview moves, no writes
sensei rebalance --apply      # Write changes to disk
sensei rebalance --cap 4      # Flag days with more than 4 reviews

Finds days exceeding the review cap and displaces the most-reviewed problems (most stable memory) to nearby low-load dates within ±50% of their current interval. Always preview before applying.

sensei new <num> <slug> <cat> — scaffold

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

sensei open <problem> — jump in

sensei open 217                    # Editor + browser
sensei open 217 --no-browser       # Editor only

sensei status — quick pulse

sensei status

Returns {total, overdue, due_today, upcoming, future, problems[]} — one-liner for agents. Schema matches sensei revisit --json counts.

---

🔧 Fuzzy Matching

Every command that takes a problem accepts:

sensei show 217                    # Problem number
sensei show contains-duplicate     # URL slug
sensei show "valid anagram"        # Title words

It matches the first unique result. No tab-complete needed (but we have it anyway — see completions).

---

🐚 Zsh Completions

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

---

📄 License

PolyForm Noncommercial License 1.0.0 — free for personal and non-commercial use.

Full license in LICENSE.

---

_{Built with precision by OverKill Labs}