# please check semantic search is working
## Retrieval Notes
- Session id: `a42467b5-6963-4806-b8cd-049dbd6ab324`
- Last activity: `2026-01-28T18:37:25.148Z`
## My Notes
<!-- Add your notes here. This section is preserved across syncs. -->
## Conversation
### User
please check semantic search is working
### Assistant
**Skill Evaluation:**
1. smart-note - NO (not creating notes)
2. google-tasks-sync - NO (not about tasks)
3. calendar-query - NO (not about calendar)
4. gmail-search - NO (not about email)
5. **semantic-search - YES** (testing this exact functionality)
6. research-query - NO
7. paper-support - NO
8-10. Reference files - NO (not needed for this test)
Let me run a test semantic search query:
### User
?
### Assistant
Hmm, that threw an error. Let me try a few more queries to diagnose whether this is a persistent issue:
### Assistant
**Result: Partial functionality**
| Function | Status | Notes |
|----------|--------|-------|
| `find_related` | ✅ Working | Found related daily notes with similarity scores |
| `semantic_search` | ❌ Error | "only length-1 arrays can be converted to Python scalars" |
| `get_context_blocks` | ❌ Error | Same error |
**Diagnosis:** The `find_related` function works (it uses pre-computed file embeddings), but query-based searches fail. This suggests the Smart Connections MCP server is having trouble generating embeddings for new query strings—likely an issue with the embedding model configuration or a version mismatch.
**Possible fixes:**
1. Restart the Smart Connections MCP server
2. Check if the embedding model in Smart Connections settings is properly configured
3. Re-index the vault in Obsidian (Smart Connections → Force Refresh)
4. Check the Smart Connections plugin is up to date
Want me to check if there's a way to restart the MCP server, or would you prefer to troubleshoot in Obsidian directly?
*Il problema sembra risiedere nella generazione degli embedding per le query, non negli indici già calcolati.*