WIP: Runtime: integrate embeddings into engine (bge-hch.5.22) #186
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Outdated suggestions cannot be applied.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.
Goal: integrate embedding-based similarity into the Director runtime scoring pipeline (thematic/lore/voice) behind an opt-in config flag.\n\nWhat I changed:\n- Added default flag in .\n- : evaluate() now optionally computes embedding similarity metrics (when enabled) and passes them into via . accepts these metrics and uses them for thematic/lore/voice placeholder slots.\n- Safe fallbacks to placeholder/default values when embeddings unavailable.\n\nTests: ran full test suite locally (unit+smoke). All tests pass.\n\nReviewer notes:\n- This change is opt-in; nothing changes unless is set in config or passed into evaluate().\n- Embedding computation uses when available. The director remains synchronous for computeRiskScore by accepting precomputed .\n\nLinked bead: ge-hch.5.22\n