Repository navigation
Separate RAG instructions and inputs into system and user messages - #5
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Task instructions were bundled with questions and source text in a user message. The pipeline now sends task rules in a system message and dynamic inputs in a JSON user message for writing, verification, repair, follow-up resolution, memory summaries and OpenUI generation.
LangSmith records the actual role-based messages. Quota estimation counts both messages, and prompt version
strict_v5_message_rolesdistinguishes new answers from previous cached results. Strict response schemas and the existing one-repair release checks are preserved.Validation passed with 154 local Python tests and 3 subtests, plus 16 separate checks in the RAGAS environment. Tests inspect the actual outbound HTTP messages, dynamic repair findings, OpenUI prompts and quota estimation. No live model calls were made, so a quality or latency improvement has not been measured.