diff --git a/README.md b/README.md
new file mode 100644
index 0000000..df7f439
--- /dev/null
+++ b/README.md
@@ -0,0 +1,7 @@
+Git clone remap repo into local project s.t. you can run code such as LLM teacher. Need an antrhopic API key, can send you mine or push a hidden version here, but don't think we should be doing it this way personally.
+
+git clone https://github.com/eric-hsiung/REMAP.git
+## TODO
+1. Get preference querying working for Kuhn Poker with perfect information game working. Currently have an HOA and TLSF file for Kuhn poker. Can use dot_trace_generator.py to generate a lot of traces of games for data gen
+2. Need to figure out if REMAP training data intake works as I think it does.
+3. Once we get reward machine LLM working we can go onto step two of L* modification