{
"event": "PreToolUse",
"tool_name": "Grep",
"tool_input": {
"pattern": "EMBEDDING_DIMENSION|EMBED.*DIM|mxbai",
"path": "\/var\/www\/scripts\/pipeline",
"output_mode": "content"
}
}
{
"tool_response": {
"mode": "content",
"numFiles": 0,
"filenames": [],
"content": "scripts\/pipeline\/__init__.py:7:- Embedding generation via Ollama (mxbai-embed-large)\nscripts\/pipeline\/model_registry.py:133: return \"ollama:mxbai-embed-large:latest\"\nscripts\/pipeline\/config.py:49:EMBEDDING_MODEL = \"mxbai-embed-large\" # 1024-dim, max ~1600 chars per chunk\nscripts\/pipeline\/config.py:50:EMBEDDING_DIMENSION = 1024\nscripts\/pipeline\/embed.py:3:Uses Ollama (mxbai-embed-large) for vector embeddings.\nscripts\/pipeline\/embed.py:12:from config import EMBEDDING_DIMENSION, EMBEDDING_MODEL, OLLAMA_HOST, QDRANT_HOST, QDRANT_PORT\nscripts\/pipeline\/embed.py:15:# Max chars for mxbai-embed model (512 token context, varies by content)\nscripts\/pipeline\/embed.py:84: if len(embedding) != EMBEDDING_DIMENSION:\nscripts\/pipeline\/embed.py:85: db.log(\"ERROR\", f\"Wrong embedding dimension: {len(embedding)} vs {EMBEDDING_DIMENSION}\")",
"numLines": 9
}
}