REST Storing - Ragrails

Three endpoints manage chunks in a vector database, the same as the SDK store(), edit(), and delete(). The collection is created automatically if it doesn’t exist.

Store (`POST /v1/store`)

Upsert embedded chunks (each with an embedding field) into the vector database.

curl -X POST http://127.0.0.1:8000/v1/store \
  -H "Content-Type: application/json" \
  -d '{
    "embedded_chunks": [
      { "id": "guide-0", "text": "...", "embedding": [0.0132, -0.0719], "metadata": {"title": "Guide"} }
    ],
    "vector_db": "qdrant",
    "collection": "docs",
    "url": "http://localhost:6333",
    "batch_size": 64
  }'

{ "inputs": 12, "stored": 12, "items": [], "failed": 0, "provider": "qdrant", "collection": "docs", "errors": [] }

Field	Default	Description
`embedded_chunks`	required	Chunks with `id`, `text`, `metadata`, `embedding`
`vector_db`	`"qdrant"`	`qdrant`, `qdrant_cloud`, `pinecone`, or `weaviate`
`collection`	`null`	Collection name
`url`	`null`	Vector DB URL
`batch_size`	`64`	Chunks per upsert request
`ensure_collection`	`true`	Create the collection if missing
`options`	`null`	Provider-specific options

Edit (`POST /v1/edit`)

Re-embed and replace stored chunks by ID. Pass the updated chunks with the same IDs; the server embeds them with the given provider before upserting.

curl -X POST http://127.0.0.1:8000/v1/edit \
  -H "Content-Type: application/json" \
  -d '{
    "chunks": [
      { "id": "guide-0", "text": "Updated text.", "source": "files/guide.pdf", "metadata": {"title": "Guide"} }
    ],
    "provider": "voyage",
    "model": "voyage-3",
    "vector_db": "qdrant",
    "collection": "docs",
    "url": "http://localhost:6333"
  }'

{ "requested": 1, "edited": 1, "items": [], "failed": 0, "provider": "qdrant", "collection": "docs", "errors": [] }

Field	Default	Description
`chunks`	required	Chunks to update (need `id` and `text`)
`provider` / `model`	`voyage` / `voyage-3`	Embedding provider used to re-embed
`vector_db` / `collection` / `url`	see store	Vector DB target
`batch_size`	`64`	Chunks per request

Delete (`POST /v1/delete`)

Remove chunks by ID.

curl -X POST http://127.0.0.1:8000/v1/delete \
  -H "Content-Type: application/json" \
  -d '{
    "ids": ["guide-0", "guide-1"],
    "vector_db": "qdrant",
    "collection": "docs",
    "url": "http://localhost:6333"
  }'

{ "requested": 2, "deleted": 2, "items": [], "failed": 0, "provider": "qdrant", "collection": "docs", "errors": [] }

Field	Default	Description
`ids`	required	Chunk IDs to delete
`vector_db` / `collection` / `url`	see store	Vector DB target

Give chunks stable IDs at chunk time so edit and delete can target them later. See Knowledge Base Maintenance.

Supported databases

Database	Notes
`qdrant`	Local or cloud. Needs `url`
`qdrant_cloud`	Cloud `url` + `QDRANT_API_KEY`
`pinecone`	Cloud. Uses `PINECONE_API_KEY`
`weaviate`	Local or cloud. Needs `url`