Introduction to CodeGraphContext

CodeGraphContext (CGC) is a high-performance, developer-focused Code Intelligence Engine that transforms source repositories into semantic, queryable property graphs. Tree-sitter and optional SCIP indexers extract symbols; CGC resolves calls, imports, and inheritance into a graph you can query from the CLI, MCP tools, or the HTTP API gateway.

Current release: 0.5.0

Recent improvements in the 0.5.0 line include cross-language call-graph resolution fixes (Perl, Ruby, Lua, Haskell, Rust, TypeScript, C#, Dart, C), structural edge persistence (PARTIAL_OF, IMPLEMENTS, METACLASS, etc.), and average CALLS audit accuracy rising from ~84% to ~98%.

Core Capabilities

Semantic AST Extraction: Utilizes tree-sitter for syntax analysis and SCIP (Sourcegraph Code Intelligence Protocol) for static symbol resolution across multiple directories.
Model Context Protocol (MCP) Integration: Built-in MCP server support allows AI models and IDE agents (Cursor, Claude, VS Code, Windsurf) to query the codebase context dynamically.
Pluggable Database Architecture: FalkorDB Lite on Unix (Python 3.12+), KuzuDB as the cross-platform fallback, plus LadybugDB, FalkorDB Remote, Nornic, and Neo4j. See configuration defaults.
Filesystem Synchronization: Integrated directory watchers monitor file updates and update the graph incrementally.
Portable Code Graphs: Supports exporting and importing serialized graph representations as .cgc bundles for offline sharing and registry integration.

Architectural Layout

CGC acts as the translation layer between source code parsing engines, graph datastores, and consumer clients.

graph TD
    src[Source Code] --> parse[Tree-sitter & SCIP Ingestion]
    parse --> builder[Graph Builder & Linker]
    builder --> db[Graph Storage: KuzuDB, FalkorDB, Neo4j]
    db --> cli[CGC CLI Client]
    db --> mcp[MCP Server Gateway]
    mcp --> ai[AI Assistant Interfaces]

Documentation Roadmap

To get started with CodeGraphContext, follow the structured sections below:

Getting Started: Explore prerequisites, installation steps, quickstart tutorials, and MCP setup.
Core Concepts: Deep dive into the architecture, graph model schemas, database backends, and parser designs.
User Guides: Learn indexing strategies, workspaces contexts, bundles distribution, custom visualizers, and database schema mappings.
Reference Manual: CLI commands, HTTP API, MCP tool schemas, and configuration variables.
Community Portal: Guidelines for contributing code, extending languages support, and the project roadmap.