Graph Endpoints

The graph endpoints expose the underlying FalkorDB property graph as serializable node/edge data. These are primarily used to power network visualizations (e.g., D3.js force graphs, Cytoscape.js) or to feed graph analytics pipelines.

Full Graph Data

GET /api/graph

Returns all nodes and edges in the graph, formatted for visualization libraries.

This endpoint returns the full graph and its payload can be several megabytes. For interactive work, use /api/graph/neighbors/{name} for a focused subgraph; use the full graph for offline processing or static visualization.

Example

curl -H "Authorization: Bearer $FL_API_TOKEN" \
  https://api.agentic.forwardlane.com/api/graph

Response Format

The values below are illustrative. Use node_count and edge_count in the response for the current graph size.

{
  "nodes": [
    {
      "id": "repo:forwardlane-backend",
      "label": "Repository",
      "properties": {
        "name": "forwardlane-backend",
        "language": "Python",
        "tier": 1,
        "score": 98.4
      }
    },
    {
      "id": "func:forwardlane-backend:validate_portfolio_allocation",
      "label": "Function",
      "properties": {
        "name": "validate_portfolio_allocation",
        "repo": "forwardlane-backend",
        "complexity": 22
      }
    }
  ],
  "edges": [
    {
      "id": "e:0",
      "source": "repo:forwardlane-backend",
      "target": "func:forwardlane-backend:validate_portfolio_allocation",
      "label": "HAS_FUNCTION"
    }
  ],
  "node_count": 22000,
  "edge_count": 75000
}

Node Object

Field	Type	Description
`id`	string	Unique node identifier in `type:name` or `type:repo:name` format.
`label`	string	Node type label (e.g. `Repository`, `Function`, `Contributor`).
`properties`	object	All properties of the node.

Edge Object

Field	Type	Description
`id`	string	Unique edge identifier.
`source`	string	Source node `id`.
`target`	string	Target node `id`.
`label`	string	Relationship type (e.g. `HAS_FUNCTION`, `DEPENDS_ON`, `COMMITTED_TO`).

Repository Neighbors

GET /api/graph/neighbors/{name}

Returns the immediate neighborhood of a repository node: all nodes directly connected to it by one hop, and the edges connecting them. This is the recommended way to build focused subgraph visualizations without loading the entire graph.

Path Parameters

Parameter	Type	Description
`name`	string	Repository name (e.g. `forwardlane-backend`).

Example

curl -H "Authorization: Bearer $FL_API_TOKEN" \
  https://api.agentic.forwardlane.com/api/graph/neighbors/forwardlane-backend

Response

{
  "center": {
    "id": "repo:forwardlane-backend",
    "label": "Repository",
    "properties": {
      "name": "forwardlane-backend",
      "language": "Python",
      "tier": 1
    }
  },
  "nodes": [
    {
      "id": "dep:django",
      "label": "Dependency",
      "properties": {"name": "django", "version": "4.2.7"}
    },
    {
      "id": "contrib:alex-chen",
      "label": "Contributor",
      "properties": {"name": "Alex Chen", "total_commits": 2341}
    }
  ],
  "edges": [
    {
      "id": "e:101",
      "source": "repo:forwardlane-backend",
      "target": "dep:django",
      "label": "DEPENDS_ON"
    },
    {
      "id": "e:102",
      "source": "contrib:alex-chen",
      "target": "repo:forwardlane-backend",
      "label": "COMMITTED_TO"
    }
  ],
  "neighbor_count": 187
}

Using Graph Data with Visualization Libraries

D3.js Force Graph

const response = await fetch(
  "https://api.agentic.forwardlane.com/api/graph/neighbors/forwardlane-backend",
  { headers: { Authorization: `Bearer ${token}` } }
);
const { nodes, edges } = await response.json();

// Transform for D3 force simulation
const d3Nodes = nodes.map(n => ({ id: n.id, label: n.label, ...n.properties }));
const d3Links = edges.map(e => ({ source: e.source, target: e.target, type: e.label }));

const simulation = d3.forceSimulation(d3Nodes)
  .force("link", d3.forceLink(d3Links).id(d => d.id))
  .force("charge", d3.forceManyBody())
  .force("center", d3.forceCenter(width / 2, height / 2));

Cytoscape.js

const { nodes, edges } = await fetchGraph();

const cytoscapeElements = [
  ...nodes.map(n => ({
    data: { id: n.id, label: n.label, ...n.properties }
  })),
  ...edges.map(e => ({
    data: { id: e.id, source: e.source, target: e.target, label: e.label }
  }))
];

const cy = cytoscape({
  container: document.getElementById("graph"),
  elements: cytoscapeElements,
  layout: { name: "cose" }
});

Graph Query Tips

Use /api/graph/neighbors/{name} for interactive repo exploration — it is far faster and lighter than the full graph.
The full /api/graph endpoint is best for offline batch processing or generating static visualizations.
Node id values are stable across API calls and can be used to join graph data with other API responses.
See Cypher Examples for complex multi-hop queries using the MCP query tool.

Get Started

Products

API Reference

CLI

FL Web

Python SDK

MCP Server

Graph Schema

Graph

Graph Endpoints

Full Graph Data

Example

Response Format

Node Object

Edge Object

Repository Neighbors

Path Parameters

Example

Response

Using Graph Data with Visualization Libraries

D3.js Force Graph

Cytoscape.js

Graph Query Tips

​Graph Endpoints

​Full Graph Data

​Example

​Response Format

​Node Object

​Edge Object

​Repository Neighbors

​Path Parameters

​Example

​Response

​Using Graph Data with Visualization Libraries

​D3.js Force Graph

​Cytoscape.js

​Graph Query Tips

Graph Endpoints

Full Graph Data

Example

Response Format

Node Object

Edge Object

Repository Neighbors

Path Parameters

Example

Response

Using Graph Data with Visualization Libraries

D3.js Force Graph

Cytoscape.js

Graph Query Tips