Skip to main content
POST
/
namespaces
/
{namespace_name}
/
items
/
get
curl -X POST "http://localhost:8080/namespaces/my-documents/items/get" \
  -H "Content-Type: application/json" \
  -d '{
    "ids": ["doc-1", "doc-2"]
  }'

{
  "status": "success",
  "message": "Successfully retrieved 1 items from namespace 'my-documents'.",
  "requested_ids": 2,
  "found_items": 1,
  "items": [
    {
      "id": "doc-1",
      "metadata": { "team": "ai" },
      "text": "Moorcheh on-prem retrieval test"
    }
  ]
}

Documentation Index

Fetch the complete documentation index at: https://docs.moorcheh.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

Fetch stored items by id within a namespace. Item ids are unique per namespace, not globally.
  • Text namespaces return id, text, and metadata for each found item
  • Vector namespaces return id and metadata only (vectors are not returned)
Ids that do not exist are omitted from items; the request still returns 200 with found_items less than requested_ids.

​
Path parameters

​
namespace_name
string
required
Namespace to read from.

​
Headers

​
Content-Type
string
required
Must be application/json

​
Body

​
ids
array
required
Non-empty array of item id strings. Maximum 100 ids per request.
curl -X POST "http://localhost:8080/namespaces/my-documents/items/get" \
  -H "Content-Type: application/json" \
  -d '{
    "ids": ["doc-1", "doc-2"]
  }'

​
Response fields

​
status
string
Request outcome. "success" when the lookup completed; "failure" on error.
​
message
string
Human-readable result or error description.
​
requested_ids
number
Number of ids sent in the request. Present on successful lookup.
​
found_items
number
Number of items returned in items. May be less than requested_ids if some ids were not found. Present on successful lookup.
​
items
array
Array of found items. Empty array when none of the requested ids exist. Present on successful lookup.
​
items[].id
string
Item id.
​
items[].text
string
Stored document text. Present for text namespace items only.
​
items[].metadata
object
Metadata stored with the item (custom fields from upload).
{
  "status": "success",
  "message": "Successfully retrieved 1 items from namespace 'my-documents'.",
  "requested_ids": 2,
  "found_items": 1,
  "items": [
    {
      "id": "doc-1",
      "metadata": { "team": "ai" },
      "text": "Moorcheh on-prem retrieval test"
    }
  ]
}

​
Important Notes

  • Maximum 100 ids per request — use multiple requests for larger batches
  • Missing ids are skipped; check found_items vs requested_ids
  • Vector embeddings are not included in the response