feat(mcp): v1.4.0 — FTS5 search, chunk truncation, FTS rebuild on startup

- Switch rah_search_nodes and rah_search_content from LIKE to FTS5 with word-split LIKE fallback
- Rebuild nodes_fts and chunks_fts indexes on server startup
- Cap chunk field at 10K chars in rah_get_nodes (adds chunk_truncated, chunk_length)
- Multi-word queries now work ("Tesla manufacturing" finds both words anywhere)
- Relevance-ranked results when FTS available

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
“BeeRad”
2026-02-10 15:26:51 +11:00
co-authored by Claude Opus 4.6
parent d1b63ea7bb
commit ec3ab74efe
5 changed files with 390 additions and 31 deletions
+3 -1
View File
@@ -53,7 +53,7 @@ Once connected, Claude will:
| `rah_get_context` | Get graph overview — stats, hub nodes, dimensions, recent activity |
| `rah_add_node` | Create a new node |
| `rah_search_nodes` | Search nodes by keyword |
| `rah_get_nodes` | Load nodes by ID |
| `rah_get_nodes` | Load nodes by ID (includes chunk + metadata) |
| `rah_update_node` | Update an existing node |
| `rah_create_edge` | Create connection between nodes |
| `rah_update_edge` | Update an edge explanation |
@@ -66,6 +66,8 @@ Once connected, Claude will:
| `rah_read_guide` | Read a guide by name |
| `rah_write_guide` | Create or update a custom guide |
| `rah_delete_guide` | Delete a custom guide |
| `rah_search_content` | Search through source content (transcripts, books, articles) |
| `rah_sqlite_query` | Execute read-only SQL queries (SELECT/WITH/PRAGMA) |
## Guides