diff --git a/README.md b/README.md
index 5ddbae7..34136e9 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 # ![Logo](images/logo.png) WAFL 0.1.0 [![Tests](https://github.com/fractalego/wafl/actions/workflows/development-tests1.yml/badge.svg)](https://github.com/fractalego/wafl/actions/workflows/development-tests1.yml)[![Docs](https://readthedocs.org/projects/wafl/badge/?version=latest)](https://wafl.readthedocs.io/en/latest/)
 
-Introductionhttps://www.youtube.com/watch?v=QKHSlBi5qok
+Introduction
 ============
 
 WAFL is a framework for personal agents. It integrates Large language models, speech recognition and text to speech.
diff --git a/todo.txt b/todo.txt
index 5f9a13b..fa7be71 100644
--- a/todo.txt
+++ b/todo.txt
@@ -1,4 +1,6 @@
-* knowledge cache does not cache the rules or facts
+* why do I need to re-initialise the retrievers after unpickling the knowledge?
+
+/* knowledge cache does not cache the rules or facts
 
 * multiple knowledge bases, one for internal facts and one for each indexed paths
 * perhaps a way to structure the prompt using <> tags. The memory items need to be distinct.