diff --git a/TODO.md b/TODO.md
index 6c6732e..ebc7577 100644
--- a/TODO.md
+++ b/TODO.md
@@ -4,6 +4,9 @@
 
 - Using talw.media's PRISM
 
+* [ ] PRISM development
+  * [ ] per-character context
+  * [ ] generate video alongside provided vox? perhaps audio/video inputting as a shim for the generator models?
 * [ ] remaining character ingestion: character context to be developed on prism
 * [ ] Vox
   * [ ] Adrian: rebuild voice with AI, experiment with prism-end stuff, elevenlabs if necessary