feat: Stricter AI Message Rendering to Prevent XSS #893
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
In the context of a chat application, we encounter diverse user inputs which makes it challenging to sanitize strings without distorting the code syntax or punctuation. In a react-based application, HTML is not rendered unless explicitly permitted. In LibreChat and in suspected cases of ChatGPT, markdown rendering is permitted in AI-generated messages.
Both these platforms only render markdown in AI messages, limiting the threat of harmful visual content generation to such messages. This might be the rationale for OpenAI not allowing for AI messages to be edited on their platform.
There are two specific instances where HTML rendering is conducted which are now secure and immune to hacky workarounds. Earlier, it was possible to bypass the security in a convoluted manner, but now, any attempts to edit a message will automatically disqualify it from the now more-narrowly-defined HTML rendering scenarios.
Change Type
Testing
Tested against specific scenarios coded for as well as with various XSS scripts.
Checklist