v0.10: /v1/rag endpoint for Router <-> Fabric platform integration#1
Merged
Conversation
New POST /v1/rag handler returns chunks (smaller, content-addressed,
ranking-weights-configurable per call) for prompt augmentation. Router
can now delegate its RAG pre-stage to Fabric instead of running embedded
pgvector + sentence-transformers.
Adds:
- fabric.chunks table (id, source_path, source_range, content, sha256,
metadata jsonb, embedding vector(768), tsv stored)
- POST /v1/rag with weights {cosine, tsvector, recency}, filters
{source_paths, type, min_score}, max_chunk_tokens
- POST /v1/chunks companion upsert endpoint for the (still-separate)
ingestion workstream
- empty-list clean response when chunks table is empty, so the
integration can be wired and smoke-tested today
Companion PR: kronaxis-router#feature/platform-integration adds
fabric_url config + the middleware splice + fallback.
3 tasks
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Adds a new POST /v1/rag endpoint so Router can delegate its embedded RAG
pre-stage to Fabric. The endpoint returns ranked chunks (smaller and
content-addressed, unlike memos) with caller-configurable ranking weights.
content_sha256, metadata jsonb, embedding vector(768), tsv stored)
{source_paths, type, min_score}, max_chunk_tokens
chunk-ingestion workstream
integration can be wired and smoke-tested today
Companion PR (Router side): Kronaxis/kronaxis-router#1
Test plan
classifier chunk at 0.79 cosine for a router-classifier query
chat completion is augmented with the retrieved chunks
Out of scope
DO NOT merge yet -- operator review.