Context confusion #22
-
|
Hi, what's caught my eye with this plugin is its cross-not capabilities. I know almost nothing about AI except as an occasional ChatGPT or Gemini user, but I'm looking for an AI plugin that can use my 6K-note vault as context for chat interactions. The functionality I'm looking for, I guess, would be to turn my vault into something like Google Notebook LM. Can VaultAI provide that functionality? |
Beta Was this translation helpful? Give feedback.
Replies: 4 comments 3 replies
-
|
VaultAI can help a lot, but it isn't a full “notebook LM” (single-step, automatic RAG of a 6K-note vault) out of the box. It has strong editor- and note-level features (cursor-aware insertion, custom prompts, file references) that let you bring relevant note content into chat. For fully automatic, trustable cross-vault retrieval over thousands of notes you’ll want a retrieval-augmented workflow also know as RAG (index + vector search) layered on top of VaultAI. |
Beta Was this translation helpful? Give feedback.
-
|
@jebeaudoin HEY! good news, found this : https://blog.google/technology/developers/file-search-gemini-api/. i will be able to integrate this. i will try and give u and update ;) |
Beta Was this translation helpful? Give feedback.
-
|
Wow. That's great. I could in no way roll my own RAG! |
Beta Was this translation helpful? Give feedback.
-
|
@jebeaudoin Hey! Great news - the RAG feature you requested is now live! 🎉 Your question about turning your vault into something like Google Notebook LM inspired us to implement full vault-wide semantic search using Google's Gemini File What's NewVaultAI now includes RAG (Retrieval Augmented Generation) that can:
How to Use It
Technical Details
You can find more details in the updated README and the RAG Implementation Thanks for the suggestion - it pushed us to build something really cool! Let me know how it works with your vault. 😊 |
Beta Was this translation helpful? Give feedback.
VaultAI can help a lot, but it isn't a full “notebook LM” (single-step, automatic RAG of a 6K-note vault) out of the box. It has strong editor- and note-level features (cursor-aware insertion, custom prompts, file references) that let you bring relevant note content into chat.
For fully automatic, trustable cross-vault retrieval over thousands of notes you’ll want a retrieval-augmented workflow also know as RAG (index + vector search) layered on top of VaultAI.