> ## Documentation Index
> Fetch the complete documentation index at: https://leetcode-py.wisl.dev/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> leetcode-py is a Python LeetCode practice environment generator with one CLI: lcpy. It is not a service or platform.
> Each problem is a directory under leetcode/ with README.md, solution.py, test_solution.py, helpers.py, and playground.ipynb. lcpy gen creates them from JSON templates bundled with the package.
> Examples are backed by tests; copy them verbatim.

# Why leetcode-py

> What a generated practice environment gives you that a bare editor does not.

Grinding LeetCode in a bare editor means writing `if __name__ == "__main__"`
prints, losing your test cases between attempts, and having nothing to show
for a solved problem. leetcode-py generates a **real software project per
problem**, tests, helpers, docs, and tooling included, so practice looks
like the engineering you already do.

## Tests come with the problem

Every generated problem ships a parametrized pytest suite with 10+ cases,
including edge cases, and a `run_helper`/`assert_helper` pair that keeps
inputs readable in failure output. You never write test scaffolding; you only
ever make red go green. Multiple solution approaches to the same problem?
One test suite covers all of them via
[parametrized classes](/practice/testing).

## Data structures you can see

`TreeNode`, `ListNode`, `GraphNode`, and `DictTree` convert between LeetCode's
array format and live objects, and render themselves as diagrams in Jupyter
or clean ASCII in the terminal. Debugging a twisted tree traversal is much
shorter when you can look at the tree.
See [Visualizations](/practice/visualizations).

## 300+ problems, ready to generate

The catalog covers the collections that matter (Grind 75, Blind 75,
NeetCode 150, NeetCode 250, and Algo Master 75) as JSON templates that
generate identical files on every machine. Browse everything in the
[Catalog](/catalog), then `lcpy gen -t <collection>` and practice.

## Extensible by design

* **Template-driven**: every problem comes from a JSON template plus a
  cookiecutter scaffold, so the tree stays consistent and reproducible.
  See [Problem Creation](/contributing/problem-creation).
* **LLM-assisted creation**: the repo ships command specs that let Claude
  Code, Cursor, or Copilot add new problems and strengthen test suites
  following the house rules. See
  [LLM-Assisted Problem Creation](/contributing/problem-creation).
* **Production hygiene**: type hints everywhere, loguru-powered test logs,
  95%+ coverage, security scanning, and CI that verifies problems regenerate
  byte-identically from their templates.

## The loop

Generate, read the README, implement `solution.py`, run `pytest`, inspect in
`playground`, next problem. Every problem, the same shape. That consistency
is the feature: nothing between you and the algorithm except the algorithm.