> For the complete documentation index, see [llms.txt](https://sopakcosauce.gitbook.io/sopakcosauce-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://sopakcosauce.gitbook.io/sopakcosauce-docs/plist-sbf-guides/whats-a-plist-sbf.md).

# What's a PList/SBF

## Short Answer

It's basically a **super compact, structured way to describe a character for an AI chatbot**, like building a “character sheet” or "profile" for your character.

## Long Answer

**PList** <mark style="background-color:red;">used to</mark> stand for **Python List,** but this method of organizing information has greatly deviated and evolved past what you typically expect to see in a Python List. Let's compare and contrast.

<table data-card-size="large" data-view="cards"><thead><tr><th></th><th data-hidden data-card-cover data-type="files"></th></tr></thead><tbody><tr><td>mylist = ["apple", "banana", "cherry"]</td><td><a href="https://3182807247-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHuZZpNB4K0yRbGFjFuhV%2Fuploads%2F8OCF9qMmfFnLeUgxjmwE%2Fprose%20(2).png?alt=media&amp;token=5c499ae1-f8a5-41bb-bbbe-d283e171ecc8">prose (2).png</a></td></tr><tr><td>[Items: apples, bananas, carrots; Notes: organic preferred]</td><td><a href="https://3182807247-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHuZZpNB4K0yRbGFjFuhV%2Fuploads%2FL1YgnBOXtfMeYhATSlfF%2Fprose%20(3).png?alt=media&amp;token=17a59965-3ef0-4a9b-ad97-93634dff7a4d">prose (3).png</a></td></tr></tbody></table>

**SBF** stands for Square Bracket Format.

So instead of writing our character "Joe Bob" in **natural language/prose** like this and using up **149 tokens**:

> Joe Bob had a face that invited trust and a smile that felt like a warm handshake. His voice was easy, casual, and always ready with a dad joke. <mark style="color:blue;">“You hear about the guy hit with a soda can? Don’t worry, it was a soft drink,”</mark> he’d quip, pausing just long enough for a groan before laughing with you, as if reminding you to lighten up.\
> \
> Always helpful without a trace of pretense, Joe Bob fixed faucets, offered rides, or left soup on a neighbor’s porch without needing thanks. <mark style="color:blue;">“Life’s tough,”</mark> he’d say with a grin, <mark style="color:blue;">“but it’s pretty funny if you look at it right.”</mark> And when he was around, it was hard not to agree.

We can reformat it to this and only use **93 tokens**:

> \[<mark style="color:orange;">Name:</mark> Joe Bob; <mark style="color:orange;">Role:</mark> community helper(jack-of-all-trades, morale booster); <mark style="color:orange;">Personality:</mark> friendly(approachable, selfless), lighthearted(jokes, optimism); <mark style="color:orange;">Speech:</mark> casual(warm, comforting), peppered with dad jokes(habit, lifts spirits); <mark style="color:orange;">Quirks:</mark> tells dad jokes(often, eases tension), fixes things for people(instinct, feels useful); <mark style="color:orange;">Psychology:</mark> warmth(expressed through humor, rooted in desire to connect)]

&#x20;It includes things like:

* The character's basic info
* Their personality traits
* How they talk and behave
* How they interact with users
* Their background story

<figure><img src="https://3182807247-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHuZZpNB4K0yRbGFjFuhV%2Fuploads%2FUWRLxLoL6OnXWUcinjwm%2F19saltines_beginner_confused_novice_scratching_head_reading_d_e5648422-cabd-41aa-b491-d803016041ce_0.png?alt=media&amp;token=4070278d-c9ae-4d0f-8d0d-c5a828867efa" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
**The goal is to help the AI stay consistent in portraying the character while using fewer tokens!**

Some creators like to use a natural, storytelling style for character descriptions instead of formats like PList/SBF or JED+. This is especially true for famous characters like Harry Potter or Satoru Gojo and common character types like tsundere or yandere.\
\
Most Large Language Models (LLMs) and Natural Language Processors (NLPs) already know these characters and types. Because of this, they can usually understand and respond to references without detailed descriptions.
{% endhint %}

***

![](https://3182807247-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHuZZpNB4K0yRbGFjFuhV%2Fuploads%2FC68qCcV0eQAB5VvyKJ7A%2Fby-nc-sa.svg?alt=media\&token=f8c998be-57a0-48ba-894b-cf2299b37a67)\ <sub>© 2024 by SopakcoSauce. Except as otherwise noted, the content of this page is licensed under</sub> [<sub>CC BY-NC-SA 4.0</sub>](https://creativecommons.org/licenses/by-nc-sa/4.0/)&#x20;