You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
docs: clarify config-driven architecture and extensibility
- Emphasize that parser is fully configurable for custom framework schemas
- Change "TypeScript/NestJS" to "TypeScript" with NestJS as included example
- Add callout that users can define domain-specific patterns and relationships
- Update feature list to highlight customizability alongside built-in NestJS support
Makes it clear the tool works with any TypeScript codebase, not just NestJS
A Model Context Protocol (MCP) server that builds rich code graphs to provide deep contextual understanding of TypeScript/NestJS codebases to Large Language Models. This server parses your codebase, constructs a comprehensive graph representation in Neo4j, and provides intelligent querying capabilities through semantic search and natural language to Cypher conversion.
7
+
A Model Context Protocol (MCP) server that builds rich code graphs to provide deep contextual understanding of TypeScript codebases to Large Language Models. This server parses your codebase using AST analysis, constructs a comprehensive graph representation in Neo4j, and provides intelligent querying capabilities through semantic search and graph traversal.
8
8
9
-
## 🌟 Features
9
+
**Config-Driven & Extensible**: Define custom framework schemas to capture domain-specific patterns beyond the included NestJS support. The parser is fully configurable to recognize your architectural patterns, decorators, and relationships.
10
10
11
-
-**📊 Rich Code Graph Generation**: Parses TypeScript/NestJS projects and creates detailed graph representations with AST-level precision
12
-
-**🔍 Semantic Search**: Vector-based semantic search using OpenAI embeddings to find relevant code patterns and implementations
13
-
-**🗣️ Natural Language Querying**: Convert natural language questions into Cypher queries using OpenAI assistants api
14
-
-**🎯 Framework-Aware**: Deep understanding of NestJS patterns (Controllers, Services, Modules, DTOs, Entities, etc.)
15
-
-**🕸️ Graph Traversal**: Explore code relationships and dependencies through intelligent graph traversal
16
-
-**⚡ High Performance**: Optimized Neo4j storage with vector indexing for fast retrieval
17
-
-**🔧 MCP Integration**: Seamless integration with Claude Code and other MCP-compatible tools
11
+
## Features
18
12
19
-
## 🏗️ Architecture
13
+
-**Rich Code Graph Generation**: Parses TypeScript projects and creates detailed graph representations with AST-level precision
14
+
-**Semantic Search**: Vector-based semantic search using OpenAI embeddings to find relevant code patterns and implementations
15
+
-**Natural Language Querying**: Convert natural language questions into Cypher queries using OpenAI assistants API
16
+
-**Framework-Aware & Customizable**: Built-in NestJS schema with ability to define custom framework patterns via configuration
17
+
-**Graph Traversal**: Explore code relationships and dependencies through intelligent graph traversal
18
+
-**High Performance**: Optimized Neo4j storage with vector indexing for fast retrieval
19
+
-**MCP Integration**: Seamless integration with Claude Code and other MCP-compatible tools
20
+
21
+
## Architecture
20
22
21
23
The MCP server consists of several key components:
22
24
@@ -34,7 +36,7 @@ The system uses a dual-schema approach:
| `traverse_from_node` | **Focused graph traversal** - Explore specific relationships from a known node | `nodeId` (string), `maxDepth?` (1-10, default:3), `skip?` (default:0) | **Deep diving** into specific code relationships. Pagination for large graphs |
502
504
| `natural_language_to_cypher` | **AI-powered query generation** - Convert natural language to Cypher queries using GPT-4 | `query` (string) | **Advanced queries** - currently requires OpenAI assistant setup |
0 commit comments