solvekit

Toolkit for extending solveit dialogs with custom toolbar buttons and dialog operations

Usage

Installation

Install latest from pypi:

$ pip install solvekit

Documentation

Documentation can be found hosted on this GitHub repository’s pages. Additionally you can find package manager specific guidelines on pypi respectively.

How to use

from solvekit import *

Add a toolbar button by calling add_btn with a name, icon, code snippet, and tooltip. Here’s the simplest case — a single-line Python snippet:

add_btn('dice', 'ii:lucide:dice-5', 'import random; dice_num = random.randint(1,6); print(f"🎲 You rolled a {dice_num}!")', 'Roll a dice!', lang='py')

For longer scripts, use a triple-quoted string. This button picks a random answer and displays it with a short delay:

add_btn('magic8', '🔮', '''
import random
import time
answers = ["It is certain! 🎉","Ask again later 🤔","Don't count on it 😬","Absolutely! ✨","Try again tomorrow 😴","Without a doubt! 🚀"]
print(f"🔮 {random.choice(answers)}")
time.sleep(5)
''', 'Ask the Magic 8-Ball', lang='py')

Buttons can also trigger AI prompts. This one asks the assistant to clean up the code cell above the current cursor position:

add_btn('cleanup', 'ii:lucide:wand-sparkles', '''
await add_msg(content="""Please clean up the code above:

1. **Fastcore docs format**: Add docstrings and a comment after each parameter (each param on its own line)
2. **General cleanup**: Improve naming, simplify logic, remove duplication

Keep the overall structure and behavior the same.""",
msg_type='prompt', run=True)
''', 'Clean up code above', lang='py')

The same pattern works for any reusable prompt — here’s one that polishes prose and formatting:

add_btn('optimize', 'ii:lucide:pencil-line', '''
await add_msg(content="""Please optimize the cell input above:

1. **Clarity**: Make the content clearer and more concise
2. **Structure**: Improve formatting and organization
3. **Grammar & Style**: Fix any grammar issues and improve readability

Keep the original intent and meaning the same.""",
msg_type='prompt', run=True)
''', 'Optimize cell input above', lang='py')

## Next steps

add_btn gives you a single toolbar button — useful, but real workflows need more. Here’s where solvekit is headed:

Dropdown menus. A single toolbar icon that expands into a list of actions, perfect for storing frequently-used prompts.

Token usage bar. A live progress bar showing context window consumption for the current session.

JS helper library. Clean wrappers around solveit internals — solvekit.run_msg(), solvekit.upsert_msg() — so custom buttons are readable and concise, no guessing at private APIs.