Index object in GraphRAG

Manually trigger GraphRAG indexing for a vault object. The object must already be ingested (completed status). This extracts entities, relationships, and communities from the document for advanced knowledge graph queries.

POST /vault/{id}/graphrag/{objectId}

cURL

cURL

curl -X POST "https://api.case.dev/vault/%3Cstring%3E/graphrag/%3Cstring%3E" \ -H "Authorization: Bearer $CASEDEV_API_KEY"

200

{
  "success": true,
  "objectId": "<string>",
  "vaultId": "<string>",
  "status": "<string>",
  "stats": {
    "entity_count": 123,
    "relationship_count": 123,
    "community_count": 123
  },
  "entities": 123,
  "relationships": 123,
  "communities": 123
}

Authorizations

Authorization · string · header · required

API key starting with sk_case_

Path Parameters

id · string · required

Vault ID

objectId · string · required

Vault object ID

Response

200

application/json

GraphRAG indexing completed successfully

success · boolean · required

Whether indexing completed successfully

objectId · string · required

ID of the indexed object

vaultId · string · required

ID of the vault

status · string · required

Status from GraphRAG service

stats · object · required

Extraction statistics

Show child attributes

entities · integer · required

Number of entities extracted

relationships · integer · required

Number of relationships extracted

communities · integer · required

Number of communities detected