llmify-decorator 0.0.2

Library for replacing function calls with LLM inference.

---

Replace a function call with LLM inference. Sends the source code and arguments to an LLM, which then predicts what the output should be.

[!CAUTION] Please, for the love of all things good and holy, do not use this in any sort of production setting. This library should only be used for experimentation or prototyping.

📦 Installation

pip install git+https://github.com/dross20/llmify

💻 Quickstart

To use llmify, simply apply it as a decorator to a function like so:

from llmify import llmify

@llmify()
def add(a, b):
  return a + b

result = add(1, 2)
print(result) # Output: 3 (probably)

To change the model used for inference, pass in a value for the model keyword argument:

@llmify(model="gpt-5")
def add(a, b):
  ...

You can also use llmify on function stubs, so long as they have docstrings or comments:

@llmify()
def greet_user(name):
  """Greet the user in a friendly manner."""
  ...

greeting = greet_user("Mortimer")
print(greeting) # Output: "Hello, Mortimer!"