@askdb/studio 0.2.0-beta.3 → 0.2.0-beta.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -17,7 +17,7 @@ It can:
17
17
  - edit table descriptions, aliases, primary entities, tags, common query language, example questions, and column metadata
18
18
  - write the describable layer back to `tables/*.md`
19
19
  - request AI enrichment suggestions when `OPENAI_API_KEY` is configured
20
- - build and query the local file-backed RAG index
20
+ - build and query the configured RAG index (`memory`, file-backed, or `pgvector`)
21
21
  - generate sample Postgres SQL for natural-language questions against the currently saved enrichment
22
22
 
23
23
  Studio uses `@askdb/enrich` for the shared non-UI Schema v2 authoring logic:
@@ -56,3 +56,7 @@ Environment variables:
56
56
  | `ASKDB_RAG_EMBEDDER` | Set to `mock`, `openai`, or `ai-sdk` for Studio RAG indexing. Defaults to the mock lexical embedder unless an AI key is configured. |
57
57
  | `ASKDB_RAG_EMBEDDER_MODEL` | Embedding model override for Studio RAG. |
58
58
  | `ASKDB_RAG_EMBEDDER_DIMENSIONS` | Optional embedding dimension override. |
59
+
60
+ Studio uses the active `rag.store` branch from `askdb.config.*`. For `pgvector`,
61
+ make sure the configured table/extension already exist and `ASKDB_PGVECTOR_URL`
62
+ resolves correctly.