FIX: keep prompt text after </document> in PromptShieldTarget parser - #2838
Open
Manohar Paturi (ManoharPaturi) wants to merge 2 commits into
Open
Manohar Paturi (ManoharPaturi) wants to merge 2 commits into
Manohar Paturi (ManoharPaturi) wants to merge 2 commits into
Conversation
The document parser sent only text before the first <document> tag as the userPrompt. Anything after a closing </document> tag, including text between two documents and the tail of the prompt, was silently dropped from the request sent to the Prompt Shield endpoint, so that part of the user prompt was never scanned for jailbreak detection. Append the post-close text to the userPrompt instead.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #2837.
the parser now appends the text after each closing tag to the userPrompt instead of discarding it, so trailing and between-document text reaches the jailbreak scan.
3 new tests cover text between documents, trailing text after the last document, and the docstring example unchanged. all fail on main and pass here. shield target suite 15/15 green, ruff clean.