Metadata-Version: 2.4
Name: llmsurgery
Version: 0.0.1
Summary: LLM Surgery
Author-email: Jeremy Howard <info@fast.ai>
License: Apache-2.0
Project-URL: Repository, https://github.com/AnswerDotAI/llmsurgery
Project-URL: Documentation, https://AnswerDotAI.github.io/llmsurgery/
Keywords: nbdev
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: fastcore>=2.0.5
Requires-Dist: python-fastllm>=0.0.28
Dynamic: license-file

# llmsurgery


<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->

## Install

``` sh
pip install llmsurgery
```

## What’s here

Dialogs are LLM conversations kept as Jupyter notebooks: notes, runnable code with outputs, and prompt/reply pairs in one editable, diffable document. This library is the data model and surgery toolkit for them, extracted from [Solveit](https://solveit.fast.ai), which subclasses these types to build its live environment:

- `llmsurgery.dialog`: [`Message`](https://AnswerDotAI.github.io/llmsurgery/dialog.html#message), [`Dialog`](https://AnswerDotAI.github.io/llmsurgery/dialog.html#dialog), and [`Attachment`](https://AnswerDotAI.github.io/llmsurgery/dialog.html#attachment), with building, navigation, sections, and merging.
- `llmsurgery.ipynb`: reading and writing dialogs as `.ipynb` files.
- `llmsurgery.hist`: converting dialogs to LLM chat history, including recovering tool calls from replies as structured messages.
- `llmsurgery.ant`: Claude Code session transcripts: read, write, curate, and build them from dialogs, so `claude --resume` opens an authored conversation.
- `llmsurgery.oai`: Codex threads: drive `codex app-server` to inject authored histories, ready for `codex resume`.

Documentation: <https://AnswerDotAI.github.io/llmsurgery/>
