Skip to main content
Supermodel provides different “lenses” to view your codebase through. Each API endpoint generates a specific type of graph suited for different analysis tasks.

​
Dependency Graph

Endpoint: /v1/graphs/dependency The Dependency Graph maps relationships at the file and module level. It answers questions like “What files import utils.ts?” or “What is the dependency structure of this module?”.
  • Nodes: Files, Modules
  • Edges: imports, exports

​
Call Graph

Endpoint: /v1/graphs/call The Call Graph provides a lower-level view, mapping function calls and class usage. It is essential for understanding execution flow, finding dead code, or tracing the impact of a function change.
  • Nodes: Functions, Classes, Methods
  • Edges: calls, instantiates, overrides

​
Domain Graph

Endpoint: /v1/graphs/domain The Domain Graph attempts to reverse-engineer the business domains and subdomains from the codebase. It clusters code artifacts into logical groups based on naming conventions, directory structure, and coupling.
  • Nodes: Domains, Subdomains
  • Edges: dependsOn, contains

​
Parse Graph

Endpoint: /v1/graphs/parse The Parse Graph exposes the raw Abstract Syntax Tree (AST) relationships. This is the most detailed view, useful for deep static analysis, linter rule creation, or refactoring tools.
  • Nodes: AST nodes (declarations, identifiers, blocks)
  • Edges: parent, child, sibling

​
Supermodel Graph

Endpoint: /v1/graphs/supermodel The Supermodel Graph is a unified representation that combines layers from all the above. It links the high-level domains down to the low-level AST nodes, providing a complete vertical slice of the system architecture.