LeetCode Tracker Agent

The LeetCode Tracker Agent automatically records problems you've solved on LeetCode, generates concise summaries, and appends them to a Google Doc while also adding flashcards to an Anki deck. It is a command‑line tool that integrates several services: LeetCode’s (unofficial) GraphQL API, the Google Docs API, OpenAI for summarisation, and AnkiConnect. The agent is modular, so you can enable or disable individual features (e.g. skip Google Docs or Anki updates) depending on your needs.

✨ Features

• Recent problem tracking – Queries LeetCode to find your most recent accepted submissions and extracts unique problems.
• Automated summarisation – Uses OpenAI’s language models to create a succinct description of each solved problem, including its difficulty and topic tags.
• Google Docs integration – Appends neatly formatted summaries to an existing Google Document, creating a running record of your progress.
• Anki flashcards – Automatically adds new notes to a specified Anki deck via AnkiConnect. Each card contains the problem title on the front and the summary with a link on the back. You can customise the note model and tags.
• Customisable – Choose how many problems to process, which models to use, how many tokens to allow in summaries, and whether to include optional tags on your flashcards.

🧰 Prerequisites

This project requires Python 3.8+. You’ll need the following external services set up:

1. LeetCode account – The tracker uses your public submission data via LeetCode’s GraphQL endpoint. No login credentials are required.
2. OpenAI API key – Required for problem summarisation. Sign up at OpenAI and set the OPENAI_API_KEY environment variable or pass --openai-key to the CLI.
3. Google Cloud credentials (optional) – To update a Google Doc you must enable the Docs API in a Google Cloud project and download a service account key JSON file with the https://www.googleapis.com/auth/documents scope.
4. Anki with AnkiConnect (optional) – Install the AnkiConnect plugin in your desktop Anki. Make sure Anki is running when you use the CLI.

Install the required Python libraries:

pip install requests openai google-api-python-client google-auth

If you plan to use the summariser or Google Docs integration, also install bs4 for HTML parsing and google-auth-oauthlib for OAuth flows:

pip install bs4 google-auth-oauthlib

🚀 Usage

python -m leetcode_tracker_agent.cli \
    --username your_leetcode_username \
    --limit 5 \
    --doc-id 1aBcDeFG2H3IjkLmNOPQ \
    --service-account-key path/to/service-account.json \
    --anki-deck "LeetCode" \
    --openai-key sk-... 

Command‑line Options

Option	Description
--username	Required. Your LeetCode username.
--limit	Number of recent solved problems to process.
--doc-id	Google Document ID to append summaries to.
--service-account-key	Path to a Google service account JSON key file. Required with --doc-id.
--anki-deck	Name of the Anki deck to add cards to.
--anki-model	Anki note model name (default: Basic).
--tags	Tags to add to all Anki notes (space‑separated).
--openai-key	Your OpenAI API key for summarisation.
--max-tokens	Maximum number of tokens for each summary (default: 400).
--temperature	Sampling temperature for summaries (default: 0.3).

Example Walkthrough

1. Enable the Google Docs API and create a service account key. Share your target document with the service account email address.
2. Start Anki on your machine and ensure the AnkiConnect plugin is installed.
3. Execute the CLI command above with appropriate arguments. The agent will:
   • Query your last limit accepted submissions on LeetCode.
   • Fetch each problem’s description via GraphQL.
   • Generate a summary using OpenAI.
   • Append the summary to the Google Doc (if configured).
   • Add a flashcard to the Anki deck (if configured).

Notes

• Data privacy: The tool sends your problem statements to the OpenAI API for summarisation. Consider whether this is acceptable for your personal or proprietary work.
• LeetCode API limits: There is no official public API; queries may break if LeetCode changes its endpoint. Use at your own risk.
• AnkiConnect must be running (i.e. Anki must be open) for note creation to succeed. If you encounter connection errors, ensure Anki is open and AnkiConnect is listening on port 8765.

📦 Project Structure

leetcode-tracker-agent/
├── leetcode_tracker_agent/    # Python package with core logic
│   ├── __init__.py
│   ├── leetcode_api.py        # Fetch solved problems via GraphQL
│   ├── summary.py             # Summarise problems using OpenAI
│   ├── google_docs.py         # Append summaries to a Google Doc
│   ├── anki_deck.py           # Add notes via AnkiConnect
│   ├── cli.py                 # Command‑line interface
│   └── types.py               # Dataclasses for problems and summaries
├── examples/                  # Placeholder for example configs or scripts
├── tests/                     # (empty) directory for future unit tests
├── README.md                  # Project overview and instructions
├── .gitignore                 # Standard gitignore for Python projects
└── LICENSE                    # MIT License

🧹 .gitignore

This repository includes a .gitignore file that excludes common Python build artefacts, virtual environments, editor settings, and Anki/Google credentials from version control. Make sure to keep any sensitive tokens or key files out of your repository.

📄 License

This project is distributed under the MIT License. You are free to use, modify, and distribute this software with attribution.

🤝 Contributing

Contributions, bug reports, and feature requests are welcome! If you have an idea to improve the agent or want to add support for additional services, please open an issue or submit a pull request.