diff --git a/README.md b/README.md
index 68cc2f4..1602cdc 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,22 @@
 # NoSlop
+
 A simple set of instructions to any LLM so it won't create slop.
+
+## What is this?
+
+NoSlop is just a collection of prompt files. There is no code, no framework, and no dependencies — only plain text prompts you can paste into any LLM to encourage clear, direct, and high-quality output.
+
+## What is "slop"?
+
+"Slop" refers to the bloated, filler-heavy, AI-generated text that large language models tend to produce by default: unnecessary caveats, hollow affirmations, redundant summaries, vague hedging, and padding that adds length without adding value.
+
+## How to use
+
+1. Browse the prompt files in this repository.
+2. Copy the contents of whichever prompt fits your use case.
+3. Paste it into the system prompt (or the beginning of your conversation) in any LLM — ChatGPT, Claude, Gemini, etc.
+4. Enjoy tighter, more useful responses.
+
+## Contributing
+
+Have a prompt that cuts through the slop? Open a pull request and add it as a new `.md` or `.txt` file in the repository.