Add DocumentDB Search: embedding-backed vector search - #1
Merged
Conversation
Adds semantic recall to the knowledge-graph search_nodes tool via a pluggable embedding provider (Ollama/local default, OpenAI, Azure OpenAI) and DocumentDB's native cosmosSearch vector index. Keyword and vector results are fused with reciprocal-rank fusion; when no embedder is configured or reachable, search degrades gracefully to keyword-only, preserving drop-in compatibility with the upstream memory server. Bundles an ollama service (plus a one-shot model pull) in compose.full.yml so the default local provider works out of the box, and adds a `graph reembed` CLI command to backfill existing entities. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 5828340b-8673-4585-9761-9b6bc4d8faca Signed-off-by: German Eichberger <geeichbe@microsoft.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 DocumentDB Search — embedding-backed semantic recall for the knowledge-graph
search_nodestool.cosmosSearchvector index.search_nodesfuses keyword ($text) and vector kNN results with reciprocal-rank fusion (RRF).@modelcontextprotocol/server-memory.compose.full.ymlnow bundles anollamaservice + one-shot model pull, so the default local provider works out of the box.documentdb-memory graph reembed [--all]CLI command backfills existing entities.Verification
cosmosSearchindex created, hybrid search returns a vector-only hit (k8s→Kubernetes) that keyword search misses.ollamahealthy, model pulled, MCP server logsembeddings: DocumentDB Search enabled(nomic-embed-text, 768 dims).Docs
README (DocumentDB Search section),
docs/architecture.md,docs/cli.md,.env.exampleall updated.