RememberIt

Python library for syncing flashcards with Anki. Designed to be easily used by LLM Agents.

Don't just solve.it.com also RememberIt!

Create beautiful styled cards with syntax highlighting, embedded images, gradient themes, and more.

Installation

pip install rememberit

📚 Tutorial

Interactive Jupyter Notebook: tutorial.ipynb

Or run the built-in tutorial:

import rememberit
rememberit.tutorial()  # Interactive walkthrough

Quick Start

import rememberit

# Login with your AnkiWeb credentials (first time only)
rememberit.login("email@example.com", "password")

# Sync and get your decks
decks = rememberit.sync()

# Create styled flashcards with code, images, and more!
deck_data = {
    "name": "Python Basics",
    "cards": [
        # Styled card (default) - random gradient theme
        {"front": "What is Python?", "back": "A programming language"},

        # Code answer with syntax highlighting
        {
            "front": "Write a function to add two numbers",
            "back": "def add(a, b):\n    return a + b",
            "back_type": "code",
        },

        # Styled question + code answer
        {
            "front": "How do you create a list comprehension?",
            "front_type": "card",
            "front_theme": "blue",
            "back": "[x**2 for x in range(10)]",
            "back_type": "code",
            "back_lang": "python",
        },

        # Images (auto-detected from paths or PIL)
        {
            "front": "System Architecture",
            "back": "~/diagrams/architecture.png",
        },
    ]
}
rememberit.upsert_deck(deck_data)

Card Schema

{
    "front": str | callable | Path | PIL.Image,  # Text, function, or image!
    "back": str | callable | Path | PIL.Image,   # Text, function, or image!
    "front_type": str,            # "card" (default) | "code" | "plain" | "image"
    "back_type": str,             # "card" (default) | "code" | "plain" | "image"
    "front_lang": str,            # For code type (default: "python")
    "back_lang": str,
    "front_theme": str,           # For card type (default: "random")
    "back_theme": str,
    "tags": str,                  # Space-separated tags
}

Card Types

Card Themes

random (default), gradient, dark, light, blue, purple, green, orange

Supported Languages

python, javascript, typescript, html, css, sql, bash, shell, json, yaml, rust, go, java, c, cpp, swift, kotlin, ruby, php, r, scala, haskell, lua, perl, markdown

Core API

Formatting

Templates

Custom templates are stored in ~/.rememberit/templates/

Utilities

Export Deck as JSON

# Get deck in upsert-compatible format
deck = rememberit.decks()["My Deck"]
deck.to_dict()  # Returns clean format with types parsed out
deck.json()     # Returns as JSON string

# Get raw HTML (old behavior)
deck.to_dict(raw=True)

Pass Functions Directly

def fibonacci(n):
    if n <= 1:
        return n
    return fibonacci(n-1) + fibonacci(n-2)

deck_data = {
    "name": "Code Examples",
    "cards": [
        {
            "front": "Fibonacci function",
            "back": fibonacci,  # Source extracted automatically!
        }
    ]
}
rememberit.upsert_deck(deck_data)

Working with Images

Images are automatically detected and converted to embedded data URIs. No manual formatting needed!

Supported Image Inputs

from pathlib import Path
from PIL import Image

# 1. File paths (relative, absolute, or with ~)
card.update(back="~/Downloads/diagram.png")
card.update(back="/absolute/path/to/image.jpg")
card.update(back="relative/path/image.png")

# 2. Path objects
image_path = Path("~/Downloads/screenshot.png").expanduser()
card.update(back=image_path)

# 3. PIL Images
pil_image = Image.open("chart.png")
card.update(back=pil_image)

# 4. Objects with _repr_png_() or _repr_jpeg_() (matplotlib, etc.)
import matplotlib.pyplot as plt
fig, ax = plt.subplots()
ax.plot([1, 2, 3], [1, 4, 9])
card.update(back=fig)  # Auto-converted!

# 5. In upsert_deck
deck_data = {
    "name": "Visual Learning",
    "cards": [
        {
            "front": "System Architecture",
            "back": "~/diagrams/architecture.png",  # Auto-detected!
        },
        {
            "front": "Performance Graph",
            "back": pil_image,  # PIL Image works too!
            "back_type": "image",  # Optional: explicit type
        }
    ]
}
rememberit.upsert_deck(deck_data)

Manual Image Formatting

If you need more control:

from rememberit import format_image

# With custom alt text and size limit
html = format_image(
    "~/Downloads/photo.jpg",
    alt="Detailed diagram",
    max_bytes=1_000_000  # 1MB limit
)
card.update(back=html)

What Works Everywhere

All these methods support images with auto-detection:

• card.update(front=image, back=image)
• deck.add_card(front=image, back=image)
• upsert_deck({"cards": [{"front": image, "back": image}]})

License

Apache-2.0