From 9e4fac496dd3fe18a70aa4a6fa9e75af8e5806e0 Mon Sep 17 00:00:00 2001
From: razor950 <razor950@gmail.com>
Date: Sat, 6 Jun 2026 13:21:28 -0400
Subject: [PATCH] made local rag needs optional (prevents having to have torch
 / sentence-transformers if you never use local embedding)

---
 conductor/tech-stack.md  |  3 ++-
 pyproject.toml           |  4 ++++
 src/rag_engine.py        | 23 +++++++++--------------
 tests/test_rag_engine.py |  9 ++++++++-
 4 files changed, 23 insertions(+), 16 deletions(-)

diff --git a/conductor/tech-stack.md b/conductor/tech-stack.md
index a241231c..7bf584bb 100644
--- a/conductor/tech-stack.md
+++ b/conductor/tech-stack.md
@@ -28,6 +28,7 @@
 - **DeepSeek-V3:** Tier 3 Worker model optimized for code implementation.
 - **DeepSeek-R1:** Specialized reasoning model for complex logical chains and "thinking" traces.
 - **Gemini Embedding 001:** Default embedding model for RAG vector store.
+- **sentence-transformers:** Optional `local-rag` extra for fully local RAG embeddings. Not part of the default install because it pulls in PyTorch.
 
 ## Configuration & Tooling
 
@@ -57,7 +58,7 @@
   - **`/api/ask` Protocol:** Non-blocking, ID-based challenge/response for synchronous HITL approvals from external contexts.
   - **`_predefined_callbacks` and `_gettable_fields`:** AppController-owned registries that the Hook API consumes to expose any App method as a `custom_callback` action.
 
-- **src/rag_engine.py:** Core RAG implementation managing the vector store lifecycle, chunking strategies (character-based and AST-aware), and multi-provider search. Integrates with **ChromaDB** for local persistence and provides a bridge for external MCP retrieval tools.
+- **src/rag_engine.py:** Core RAG implementation managing the vector store lifecycle, chunking strategies (character-based and AST-aware), and multi-provider search. Integrates with **ChromaDB** for local persistence, uses external embeddings by default, and provides an optional local embedding path via `manual_slop[local-rag]`.
 
 - **src/beads_client.py:** Python client for interacting with the [Beads](https://github.com/steveyegge/beads) / Dolt backend. Handles repository initialization, bead creation, status updates, and graph queries.
 
diff --git a/pyproject.toml b/pyproject.toml
index b5ffafb6..52d933c1 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -24,6 +24,10 @@ dependencies = [
     "openai",
 
     "chromadb>=1.5.8",
+]
+
+[project.optional-dependencies]
+local-rag = [
     "sentence-transformers>=5.4.1",
 ]
 
diff --git a/src/rag_engine.py b/src/rag_engine.py
index 3287f1bd..08c01866 100644
--- a/src/rag_engine.py
+++ b/src/rag_engine.py
@@ -16,6 +16,7 @@ from src.file_cache import ASTParser
 _SENTENCE_TRANSFORMERS = None
 _GOOGLE_GENAI          = None
 _CHROMADB              = None
+LOCAL_RAG_INSTALL_HINT = "Local RAG embeddings require sentence-transformers. Install with manual_slop[local-rag] to use local embeddings."
 
 
 def _get_sentence_transformers():
@@ -24,6 +25,10 @@ def _get_sentence_transformers():
   try:
    from sentence_transformers import SentenceTransformer
    _SENTENCE_TRANSFORMERS = SentenceTransformer
+  except ModuleNotFoundError as e:
+   if e.name == "sentence_transformers":
+    raise ImportError(LOCAL_RAG_INSTALL_HINT) from e
+   raise
   except Exception as e:
    sys.stderr.write(f"FAILED to import sentence_transformers: {e}\n")
    sys.stderr.flush()
@@ -52,22 +57,12 @@ class BaseEmbeddingProvider:
 
 class LocalEmbeddingProvider(BaseEmbeddingProvider):
  def __init__(self, model_name: str = 'all-MiniLM-L6-v2'):
-  self.model = None
-  try:
-   ST = _get_sentence_transformers()
-   if ST:
-    self.model = ST(model_name)
-  except Exception as e:
-   sys.stderr.write(f"LocalEmbeddingProvider failed to load model {model_name}: {e}. Using dummy embeddings.\n")
-   sys.stderr.flush()
+  ST = _get_sentence_transformers()
+  self.model = ST(model_name)
 
  def embed(self, texts: List[str]) -> List[List[float]]:
-  if self.model:
-   embeddings = self.model.encode(texts)
-   return embeddings.tolist()
-  else:
-   # Dummy embeddings (384 dims for all-MiniLM-L6-v2)
-   return [[0.0] * 384 for _ in texts]
+  embeddings = self.model.encode(texts)
+  return embeddings.tolist()
 
 class GeminiEmbeddingProvider(BaseEmbeddingProvider):
  def __init__(self, model_name: str = 'gemini-embedding-001'):
diff --git a/tests/test_rag_engine.py b/tests/test_rag_engine.py
index c9d50133..370d20d5 100644
--- a/tests/test_rag_engine.py
+++ b/tests/test_rag_engine.py
@@ -2,6 +2,7 @@ import pytest
 import os
 from unittest.mock import MagicMock, patch
 from src import models
+from src import rag_engine
 from src.rag_engine import RAGEngine, BaseEmbeddingProvider, LocalEmbeddingProvider, GeminiEmbeddingProvider
 
 class MockEmbeddingProvider(BaseEmbeddingProvider):
@@ -11,13 +12,19 @@ class MockEmbeddingProvider(BaseEmbeddingProvider):
 @pytest.fixture
 def mock_rag_config():
  vs_config = models.VectorStoreConfig(provider='mock', collection_name='test')
- return models.RAGConfig(enabled=True, vector_store=vs_config, embedding_provider='local')
+ return models.RAGConfig(enabled=True, vector_store=vs_config, embedding_provider='gemini')
 
 def test_rag_engine_init_mock(mock_rag_config):
  engine = RAGEngine(mock_rag_config)
  assert engine.config.enabled is True
  assert engine.collection == "mock"
 
+def test_local_embedding_provider_missing_dependency_has_install_hint():
+ with patch.object(rag_engine, "_SENTENCE_TRANSFORMERS", None):
+  with patch.dict("sys.modules", {"sentence_transformers": None}):
+   with pytest.raises(ImportError, match=r"manual_slop\[local-rag\]"):
+    LocalEmbeddingProvider()
+
 @patch('src.rag_engine.LocalEmbeddingProvider.embed')
 @patch('src.rag_engine._get_chromadb')
 def test_rag_engine_chroma(mock_get_chroma, mock_embed):