code-graph-builder 0.21.0

Code knowledge graph builder with MCP server for AI-assisted code navigation

Code Graph Builder

Build a knowledge graph from any codebase, generate API documentation, and search code semantically — all accessible as an MCP server for AI coding assistants.

What It Does

Your Code Repository
    |
    v
[Tree-sitter AST Parsing]  ──>  Knowledge Graph (Kuzu)
    |                               |
    |                               v
    |                        API Documentation (Markdown)
    |                               |
    |                               v
    |                        Vector Embeddings
    |                               |
    v                               v
MCP Server  <──────────────  Semantic Search
    |
    v
Claude Code / Cursor / Windsurf / Any MCP Client

Core workflow for AI agents:

initialize_repository  →  find_api  →  get_api_doc

1. Index the codebase once
2. Search by vague semantic description ("PWM duty cycle update")
3. Get precise function signatures, call trees, and usage examples

Quick Start

Install via npx (recommended)

# First run — interactive setup wizard
npx code-graph-builder@latest --setup

# Start MCP server
npx code-graph-builder@latest --server

The setup wizard:

1. Auto-installs the Python package if not found
2. Configures workspace, LLM provider, and embedding provider
3. Runs an MCP smoke test to verify the server works
4. Optionally registers as a global MCP server for Claude Code (claude mcp add --scope user)

Install via pip

pip install "code-graph-builder[treesitter-c,semantic]"
cgb-mcp  # Start MCP server

Uninstall

npx code-graph-builder@latest --uninstall

Removes: Claude MCP registration, Python package, workspace data.

MCP Client Configuration

Add to your MCP client config (Claude Code, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "code-graph-builder": {
      "command": "npx",
      "args": ["-y", "code-graph-builder@latest", "--server"]
    }
  }
}

On Windows, use:

{
  "mcpServers": {
    "code-graph-builder": {
      "command": "cmd",
      "args": ["/c", "npx", "-y", "code-graph-builder@latest", "--server"]
    }
  }
}

Pipeline

Step	What	Input	Output
1. graph-build	Tree-sitter AST parsing	Source code	Kuzu graph database
2. api-doc-gen	Query graph, render docs	Graph	3-level Markdown (index / module / function)
2b. desc-gen	LLM generates descriptions	Functions without docstrings	Descriptions in L3 Markdown
3. embed-gen	Vectorize function docs	L3 Markdown files	Vector store (pickle)

Steps 1-3 run automatically via initialize_repository. Wiki generation is available separately via generate_wiki.

initialize_repository  →  Steps 1-3 (full pipeline)
build_graph            →  Step 1 only
generate_api_docs      →  Step 2 + 2b (modes: full / resume / enhance)
rebuild_embeddings     →  Step 3
generate_wiki          →  Separate (not in main pipeline)

API Doc Generation Modes

Mode	Behavior
full	Rebuild all docs from graph
resume	Generate only for functions with TODO placeholders
enhance	LLM-powered module summaries + API usage workflows

MCP Tools

Primary Tools (10 exposed)

Repository Management

Tool	Description
initialize_repository	Index a repo: graph + API docs + embeddings
get_repository_info	Active repo stats (node/relationship counts, service status)
list_repositories	All indexed repos with pipeline completion status
switch_repository	Switch active repo for queries
link_repository	Reuse existing index for a different repo path (no re-indexing)

Code Search & Documentation

Tool	Description
find_api	Semantic search + API doc (primary search tool)
list_api_docs	Browse L1 module index or L2 module details
get_api_doc	L3 function detail: signature, call tree, usage examples, source
generate_api_docs	Generate/update API docs (full / resume / enhance)
get_config	Show server configuration and service availability

Hidden Tools (11 available via handler)

These tools are superseded by the API-doc-based workflow above but remain accessible: query_code_graph, get_code_snippet, semantic_search, locate_function, list_api_interfaces, list_wiki_pages, get_wiki_page, generate_wiki, rebuild_embeddings, build_graph, prepare_guidance

API Documentation Format

Generated docs are optimized for both AI agent reading and vector retrieval.

L3 Function Detail (embedding unit)

# parse_btype

> Parse base type declaration including struct/union/enum specifiers.

- Signature: `int parse_btype(CType *type, AttributeDef *ad, int ignore_label)`
- Return: `int`
- Visibility: static | Header: tccgen.h
- Location: tccgen.c:139-280
- Module: tinycc.tccgen — C code generator

## Call Tree

parse_btype
├── expr_const           [static]
├── parse_btype_qualify   [static]
├── struct_decl           [static]
│   ├── expect
│   └── next
└── parse_attribute       [static]

## Called by (5)

- type_decl (tinycc.tccgen) → tccgen.c:1200
- post_type (tinycc.tccgen) → tccgen.c:1350

## Parameters & Memory

| Parameter | Direction | Ownership |
|-----------|-----------|-----------|
| `CType *type` | in/out | borrowed, modified |
| `AttributeDef *ad` | in/out | borrowed, modified |

## Implementation

​```c
int parse_btype(CType *type, AttributeDef *ad, int ignore_label) {
    // ... source code embedded
}
​```

C/C++ Specific Features

• Extracts // and /* */ comments above functions as descriptions
• Struct/union/enum members displayed with types
• Macro definitions in dedicated section
• Static/public/extern visibility classification
• Memory ownership inference from signatures
• Header/implementation file split
• Cross-file function call resolution via #include header mapping
• GB2312/GBK encoding support for source files

Supported Languages

Language	Functions	Classes/Structs	Calls	Imports	Types
C / C++	Yes	struct, union, enum, typedef, macro	Yes	#include	Yes
Python	Yes	Yes	Yes	Yes	-
JavaScript / TypeScript	Yes	Yes	Yes	Yes	-
Rust	Yes	struct, enum, trait, impl	Yes	Yes	-
Go	Yes	struct, interface	Yes	Yes	-
Java	Yes	class, interface, enum	Yes	Yes	-
Scala	Yes	class, object	Yes	Yes	-
C#	Yes	class, namespace	Yes	-	-
PHP	Yes	class	Yes	-	-
Lua	Yes	-	Yes	-	-

Graph Schema

Nodes: Project, Package, Module, File, Folder, Class, Function, Method, Type, Enum, Union

Relationships: CONTAINS_*, DEFINES, DEFINES_METHOD, CALLS, INHERITS, IMPLEMENTS, IMPORTS, OVERRIDES

Properties: qualified_name (PK), name, path, start_line, end_line, signature, return_type, visibility, parameters, kind, docstring

Environment Variables

LLM (first match wins)

Variable	Purpose	Default
LLM_API_KEY	Generic LLM key (highest priority)	-
LLM_BASE_URL	API endpoint	https://api.openai.com/v1
LLM_MODEL	Model name	gpt-4o
OPENAI_API_KEY	OpenAI or compatible	-
MOONSHOT_API_KEY	Moonshot / Kimi (legacy)	-

Embedding

Variable	Purpose	Default
DASHSCOPE_API_KEY	DashScope (Qwen3 Embedding)	-
DASHSCOPE_BASE_URL	DashScope endpoint	https://dashscope.aliyuncs.com/api/v1

System

Variable	Purpose	Default
CGB_WORKSPACE	Workspace directory	~/.code-graph-builder

Installation Options

# Core only (graph building)
pip install code-graph-builder

# With C/C++ support
pip install "code-graph-builder[treesitter-c]"

# With all languages
pip install "code-graph-builder[treesitter-full]"

# With semantic search
pip install "code-graph-builder[semantic]"

# Everything
pip install "code-graph-builder[treesitter-full,semantic,rag]"

Development

git clone https://github.com/JeremyJiao01/CodeGraphWiki.git
cd CodeGraphWiki
pip install -e ".[treesitter-full,semantic,rag]"

# Run tests
python3 -m pytest code_graph_builder/tests/ -v

# Integration tests (requires tinycc repo at ../tinycc)
python3 -m pytest code_graph_builder/tests/test_step1_graph_build.py -v   # ~3 min
python3 -m pytest code_graph_builder/tests/test_step2_api_docs.py -v      # ~3 min
python3 -m pytest code_graph_builder/tests/test_step3_embedding.py -v     # ~27 min (API calls)
python3 -m pytest code_graph_builder/tests/test_api_find_integration.py -v # ~47 min (full pipeline)

License

MIT