> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs-beta.getzep.com/v3/sdk-reference/graph/node/get-neighbors/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs-beta.getzep.com/_mcp/server.

# Get Node Neighbors

POST https://api.getzep.com/api/v2/graph/node/{node_uuid}/neighbors
Content-Type: application/json

Enumerates the distinct entity nodes directly connected to a node, together with the edges connecting each to it.

Reference: https://docs-beta.getzep.com/sdk-reference/graph/node/get-neighbors

## Request

### Path parameters

- `node_uuid` (string, required) — Node UUID

### Body (application/json)

This endpoint expects an apidata.GraphNodeNeighborsRequest.

- `cursor` (string, optional) — Opaque cursor for pagination, obtained from the Zep-Next-Cursor response header of the previous page.
- `direction` (string, optional) — Orientation of the connecting edge relative to the anchor node: "out" (anchor is the edge's source), "in" (anchor is the edge's target), or "both" (either). Defaults to "both".
- `direction_sort` (string, optional) — Sort direction for order_by. One of "asc" or "desc". Defaults to "desc". Named direction_sort to avoid clashing with the traversal Direction field above.
- `filters` (graphiti.SearchFilters, optional) — Filters constraining the connecting edges (edge types, dates, and the node- and episode-anchored UUID fields) and the neighbor nodes (node_labels/exclude_node_labels). Reuses the graph.search filter type.
- `limit` (integer, optional) — Maximum number of neighbor nodes to return. An explicit value is clamped to 50; when omitted, the default page size (100) applies.
- `order_by` (string, optional) — Field to sort neighbor nodes by. One of "uuid" or "created_at". Defaults to "uuid".

## Response

### 200

Neighbors

- `list of apidata.GraphNodeNeighbor`

## Errors

### 400 Bad Request Error

Bad Request

- `message` (string, optional)

### 404 Not Found Error

Not Found

- `message` (string, optional)

### 500 Internal Server Error

Internal Server Error

- `message` (string, optional)

## Types

### graphiti.SearchFilters

- `connected_node_uuids` (list of string, optional) — List of node UUIDs to filter edges on: an edge matches if its source OR target node UUID is in this list. Applies to edges only; rejected on requests whose result type contains no edges. Max 256 entries.
- `created_at` (list of list of graphiti.DateFilter, optional) — 2D array of date filters for the created\_at field. The outer array elements are combined with OR logic. The inner array elements are combined with AND logic. Example: `[[{">", date1}, {"<", date2}], [{"=", date3}]]` This translates to: `(created_at > date1 AND created_at < date2) OR (created_at = date3)`
- `edge_types` (list of string, optional) — List of edge types to filter on
- `edge_uuids` (list of string, optional) — List of edge UUIDs to filter on. Max 256 to align with graph-service filter limits.
- `episode_metadata_filters` (graphiti.MetadataFilterGroup, optional) — [Experimental] Episode metadata filter. Restricts results to edges/nodes derived from episodes matching the metadata predicates. Uses explicit AND/OR groups. This feature is experimental and may change in future releases.
- `episode_uuids` (list of string, optional) — List of episode UUIDs to filter on. An edge matches if it was derived from any listed episode; a node matches if it is mentioned by any listed episode. Valid for both edge and node result types. Max 256 entries.
- `exclude_edge_types` (list of string, optional) — List of edge types to exclude from results
- `exclude_node_labels` (list of string, optional) — List of node labels to exclude from results
- `expired_at` (list of list of graphiti.DateFilter, optional) — 2D array of date filters for the expired\_at field. The outer array elements are combined with OR logic. The inner array elements are combined with AND logic. Example: `[[{">", date1}, {"<", date2}], [{"=", date3}]]` This translates to: `(expired_at > date1 AND expired_at < date2) OR (expired_at = date3)`
- `invalid_at` (list of list of graphiti.DateFilter, optional) — 2D array of date filters for the invalid\_at field. The outer array elements are combined with OR logic. The inner array elements are combined with AND logic. Example: `[[{">", date1}, {"<", date2}], [{"=", date3}]]` This translates to: `(invalid_at > date1 AND invalid_at < date2) OR (invalid_at = date3)`
- `node_labels` (list of string, optional) — List of node labels to filter on
- `property_filters` (list of graphiti.PropertyFilter, optional) — List of property filters to apply to nodes and edges
- `source_node_uuids` (list of string, optional) — List of node UUIDs to filter edges on: an edge matches if its source node UUID is in this list. Applies to edges only; rejected on requests whose result type contains no edges. Max 256 entries.
- `target_node_uuids` (list of string, optional) — List of node UUIDs to filter edges on: an edge matches if its target node UUID is in this list. Applies to edges only; rejected on requests whose result type contains no edges. Max 256 entries.
- `valid_at` (list of list of graphiti.DateFilter, optional) — 2D array of date filters for the valid\_at field. The outer array elements are combined with OR logic. The inner array elements are combined with AND logic. Example: `[[{">", date1}, {"<", date2}], [{"=", date3}]]` This translates to: `(valid_at > date1 AND valid_at < date2) OR (valid_at = date3)`

### apidata.GraphNodeNeighbor

- `edges` (list of graphiti.EntityEdge, optional)
- `node` (graphiti.EntityNode, optional)

### graphiti.DateFilter

- `comparison_operator` (enum, required) — Comparison operator for date filter
  - Allowed values: `=`, `<>`, `>`, `<`, `>=`, `<=`, `IS NULL`, `is_null`, `IS NOT NULL`, `CONTAINS`
- `date` (string, optional) — Date to filter on. Required for non-null operators (`=`, `<>`, `>`, `<`, `>=`, `<=`). Should be omitted for IS NULL (or is\_null) and IS NOT NULL operators.

### graphiti.MetadataFilterGroup

- `type` (enum, required) — Logical operator: "and" or "or"
  - Allowed values: `and`, `or`
- `filters` (list of graphiti.EpisodeMetadataFilter, optional) — Leaf filters (predicates on metadata key-value pairs)
- `groups` (list of graphiti.MetadataFilterGroup, optional) — Nested sub-groups for composing complex boolean expressions

### graphiti.PropertyFilter

- `comparison_operator` (enum, required) — Comparison operator for property filter
  - Allowed values: `=`, `<>`, `>`, `<`, `>=`, `<=`, `IS NULL`, `is_null`, `IS NOT NULL`, `CONTAINS`
- `property_name` (string, required) — Property name to filter on
- `property_value` (any, optional) — Property value to match on. Accepted types: string, int, float64, bool, or nil. Invalid types (e.g., arrays, objects) will be rejected by validation. Must be non-nil for non-null operators (`=`, `<>`, `>`, `<`, `>=`, `<=`).

### graphiti.EntityEdge

- `created_at` (string, required) — Creation time of the edge
- `fact` (string, required) — Fact representing the edge and nodes that it connects
- `name` (string, required) — Name of the edge, relation name
- `source_node_uuid` (string, required) — UUID of the source node
- `target_node_uuid` (string, required) — UUID of the target node
- `uuid` (string, required) — UUID of the edge
- `attributes` (map from string to any, optional) — Additional attributes of the edge. Dependent on edge types
- `episodes` (list of string, optional) — List of episode ids that reference these entity edges
- `expired_at` (string, optional) — Datetime of when the node was invalidated
- `hyperedge_uuid` (string, optional) — HyperedgeUUID groups the pairwise edges projected from the same atomic multi-entity fact. Omitted when the edge is not part of a hyperedge.
- `invalid_at` (string, optional) — Datetime of when the fact stopped being true
- `relevance` (double, optional) — Relevance is an experimental rank-aligned score in [0,1] derived from Score via logit transformation. Only populated when using cross_encoder reranker; omitted for other reranker types (e.g., RRF).
- `scope` (string, optional) — Scope of the edge (e.g. "entity", "maybe_related")
- `score` (double, optional) — Score is the reranker output: sigmoid-distributed logits [0,1] when using cross_encoder reranker, or RRF ordinal rank when using rrf reranker
- `selection_rank` (integer, optional) — SelectionRank is the global cross-scope rank assigned by auto scope selection.
- `source_node_labels` (list of string, optional) — SourceNodeLabels are the labels of the source node at read time. Same read-time-projection semantics as SourceNodeName.
- `source_node_name` (string, optional) — SourceNodeName is the name of the source node at read time. It is a read-time projection of current node state, not a stored edge attribute: a subsequent node rename is reflected on the next read. Omitted (the edge is still returned) if the source node cannot be resolved, for example if it was deleted concurrently.
- `target_node_labels` (list of string, optional) — TargetNodeLabels are the labels of the target node at read time. Same read-time-projection semantics as SourceNodeName.
- `target_node_name` (string, optional) — TargetNodeName is the name of the target node at read time. Same read-time-projection semantics as SourceNodeName.
- `valid_at` (string, optional) — Datetime of when the fact became true

### graphiti.EntityNode

- `created_at` (string, required) — Creation time of the node
- `name` (string, required) — Name of the node
- `summary` (string, required) — Regional summary of surrounding edges
- `uuid` (string, required) — UUID of the node
- `attributes` (map from string to any, optional) — Additional attributes of the node. Dependent on node labels
- `degree` (integer, optional) — The count of live entity edges that touch this node (the edge scopes the entity edge list returns). Present only when the list request orders by degree.
- `episodes` (list of string, optional) — The UUIDs of the live episodes that mention this node, newest first. The list is complete when `episodes_truncated` is false. The list is empty when the node has more than 100 source episodes; list episodes with the `mentioned_node_uuids` filter to read them.
- `episodes_truncated` (boolean, optional) — True when the node has more than 100 source episodes, so `episodes` is empty, or when provenance is unavailable. False means `episodes` is the complete set.
- `labels` (list of string, optional) — Labels associated with the node
- `relevance` (double, optional) — Relevance is an experimental rank-aligned score in [0,1] derived from Score via logit transformation. Only populated when using cross_encoder reranker; omitted for other reranker types (e.g., RRF).
- `score` (double, optional) — Score is the reranker output: sigmoid-distributed logits [0,1] when using cross_encoder reranker, or RRF ordinal rank when using rrf reranker
- `selection_rank` (integer, optional) — SelectionRank is the global cross-scope rank assigned by auto scope selection.

### graphiti.EpisodeMetadataFilter

- `comparison_operator` (enum, required) — Comparison operator: =, \<>, >, \<, >=, \<=, IS NULL, IS NOT NULL, IN, CONTAINS
  - Allowed values: `=`, `<>`, `>`, `<`, `>=`, `<=`, `IS NULL`, `is_null`, `IS NOT NULL`, `CONTAINS`
- `property_name` (string, required) — Metadata key to filter on
- `property_value` (any, optional) — Value to compare against. Not required for IS NULL / IS NOT NULL operators.

## Examples

**Request**

```json
{}
```

**Response**

```json
[
  {
    "edges": [
      {
        "created_at": "string",
        "fact": "string",
        "name": "string",
        "source_node_uuid": "string",
        "target_node_uuid": "string",
        "uuid": "string",
        "attributes": {},
        "episodes": [
          "string"
        ],
        "expired_at": "string",
        "hyperedge_uuid": "string",
        "invalid_at": "string",
        "relevance": 1.1,
        "scope": "string",
        "score": 1.1,
        "selection_rank": 1,
        "source_node_labels": [
          "string"
        ],
        "source_node_name": "string",
        "target_node_labels": [
          "string"
        ],
        "target_node_name": "string",
        "valid_at": "string"
      }
    ],
    "node": {
      "created_at": "string",
      "name": "string",
      "summary": "string",
      "uuid": "string",
      "attributes": {},
      "degree": 12,
      "episodes": [
        "string"
      ],
      "episodes_truncated": true,
      "labels": [
        "string"
      ],
      "relevance": 1.1,
      "score": 1.1,
      "selection_rank": 1
    }
  }
]
```