diff --git a/README.md b/README.md
new file mode 100644
index 0000000..d473d73
--- /dev/null
+++ b/README.md
@@ -0,0 +1,46 @@
+# Dochunt
+
+An OSINT tool to run public vk documents though an OCR system.
+
+## Install:
+```sh
+git clone https://github.com/anatolykopyl/dochunt.git
+cd dochunt
+pip install -r requirements.txt
+```
+
+## Usage:
+
+1. Set up your credentials in `config.ini`.
+
+2. Set up whatever you're interested in in `config.ini` 
+in `interests`. This should be either a list of strings
+formatted like this:
+```ini
+interests = [
+    "interest1",
+    "interest2",
+    "interest3",
+    ...
+  ]
+```
+If the image contains any of these strings the script will save it.
+
+Or the word `any` to save any image containing text.
+```ini
+interests = any
+```
+
+3. Run the script 
+
+This will watch the latest uploaded document:
+```sh
+python main.py
+```
+This will go through all availible documents (vk caps search to 1000 latest docs):
+```sh
+python main.py -a
+```
+
+
+Inspired by [darkshot](https://github.com/mxrch/darkshot).
\ No newline at end of file