Graph Relations By Source ID

Graph Relations By Id

import requests

url = "https://api.hydradb.com/list/graph_relations_by_id"

headers = {"Authorization": "Bearer <token>"}

response = requests.get(url, headers=headers)

print(response.text)

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.hydradb.com/list/graph_relations_by_id', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

{
  "relations": [],
  "success": true,
  "message": "Relations retrieved successfully"
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

GET

list

graph_relations_by_id

Graph Relations By Id

import requests

url = "https://api.hydradb.com/list/graph_relations_by_id"

headers = {"Authorization": "Bearer <token>"}

response = requests.get(url, headers=headers)

print(response.text)

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.hydradb.com/list/graph_relations_by_id', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

{
  "relations": [],
  "success": true,
  "message": "Relations retrieved successfully"
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

{
  "detail": {
    "success": true,
    "message": "Error occurred",
    "error_code": "<string>"
  }
}

When to use it

Debugging graph construction – verify HydraDB extracted the relationships you expected
Visualizations – build entity graphs in admin UIs
Audit – confirm what entities and connections exist for a specific document

This endpoint returns the raw triplets (subject → predicate → object) that HydraDB extracted from a single source.

Endpoint

Auth: Bearer token
Idempotency: Read-only
Async: No

Example

curl 'https://api.hydradb.com/list/graph_relations_by_id?source_id=doc_12345&tenant_id=my_first_tenant&limit=100' \
  -H "Authorization: Bearer <your_api_key>"

const response = await client.fetch.graphRelationsBySourceId({
  source_id: "doc_12345",
  tenant_id: "my_first_tenant",
  limit: 100
});

response = client.fetch.graph_relations_by_source_id(
    source_id="doc_12345",
    tenant_id="my_first_tenant",
    limit=100,
)

Query parameters

Name	Type	Required	Default	Description
`source_id`	string	No (defaults to root)	–	The source to fetch relations for. Optional - if omitted, returns top-level entity relations.
`tenant_id`	string	Yes	–	The tenant the source belongs to.
`sub_tenant_id`	string	No	default	Sub-tenant scope.
`is_memory`	boolean	No	`false`	Set `true` to fetch relations for a memory; `false` for knowledge.
`limit`	integer	No	`250`	Maximum number of relations to return.

Response

{
  "success": true,
  "message": "Relations retrieved successfully",
  "relations": [
    {
      "source": {
        "name": "Pricing Strategy",
        "type": "PROJECT",
        "entity_id": "ent_abc123",
        "namespace": "default"
      },
      "target": {
        "name": "Product Team",
        "type": "ORGANIZATION",
        "entity_id": "ent_def456",
        "namespace": "default"
      },
      "relations": [
        {
          "canonical_predicate": "OWNED_BY",
          "raw_predicate": "is owned by",
          "context": "The Q4 pricing strategy was developed and is owned by the Product Team...",
          "confidence": 0.92,
          "relationship_id": "rel_xyz789",
          "chunk_id": "chunk_001"
        }
      ],
      "chunk_id": "chunk_001"
    }
  ]
}

Field	Description
`relations[]`	Array of triplets extracted from this source.
`relations[].source`	Subject entity.
`relations[].target`	Object entity.
`relations[].relations[]`	Evidence for the relationship. Multiple entries mean the same connection appears in multiple chunks.
`relations[].relations[].canonical_predicate`	Normalized relationship type (e.g., `OWNED_BY`, `REPORTS_TO`, `WORKS_FOR`).
`relations[].relations[].raw_predicate`	The original phrasing extracted from the text.
`relations[].relations[].context`	Surrounding context for the relationship.
`relations[].relations[].confidence`	Extraction confidence (0.0–1.0).
`relations[].chunk_id`	The chunk this triplet came from.

Entity types

The type field on each entity is normalized at extraction. Common types:

Type	Examples
`PERSON`	Named individuals
`ORGANIZATION`	Companies, teams, departments
`PROJECT`	Initiatives, products, releases
`PRODUCT`	Specific products or services
`ERROR_CODE`	Error identifiers
`LOCATION`	Geographic entities

The full list depends on what HydraDB extracted from your specific content.

Behavior notes

No relations doesn’t mean no graph. A source might be successfully indexed without producing extractable triplets (very short content, structured data, or content without clear entity-relationship language). Use this endpoint to confirm what was actually extracted.

Memories vs knowledge. Set is_memory: true to look up relations for a memory item. The default (false) targets knowledge sources.

Browse sources first: List data – find the source_id you want to inspect
Retrieve content: Fetch content – get the original file or text
Use in recall: Full recall with graph_context: true – see graph context applied to a query

Errors

Common codes: 400 INVALID_PARAMETERS, 404 SOURCE_NOT_FOUND, 422 VALIDATION_ERROR. See Error Responses for the full list. Read more: Essentials → Context Graphs

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Query Parameters

source_id

string

required

The source ID to fetch relations for

Example:

"<str>"

tenant_id

string

default:""

Unique identifier for the tenant/organization

Example:

"tenant_1234"

is_memory

boolean

default:false

Whether to fetch relations for memories

Example:

"<str>"

sub_tenant_id

string | null

Optional sub-tenant identifier used to organize data within a tenant. If omitted, the default sub-tenant created during tenant setup will be used.

Example:

"sub_tenant_4567"

limit

integer

default:250

Maximum number of relations to return

Example:

"<str>"

Response

Successful Response

relations

(TripletWithEvidence · object | null)[]

required

List of relations retrieved

Triple with multiple evidence items from different chunks

Show child attributes

Example:

[]

success

boolean

default:true

Indicates whether the request was successful

Example:

true

message

string

default:Relations retrieved successfully

Response message describing the operation result

API Documentation

Tenants

Knowledge & Memories

Recall

Managing Data

Graph Relations By Source ID

When to use it

Endpoint

Example

Query parameters

Response

Entity types

Behavior notes

Errors

Authorizations

Query Parameters

Response

​When to use it

​Endpoint

​Example

​Query parameters

​Response

​Entity types

​Behavior notes

​Related endpoints

​Errors

Authorizations

Query Parameters

Response

When to use it

Endpoint

Example

Query parameters

Response

Entity types

Behavior notes

Related endpoints

Errors