### Install Pandas and List Files

Source: https://github.com/tirth8205/code-review-graph/blob/main/tests/fixtures/sample_notebook.ipynb

Installs the pandas library and lists files in the current directory. Use this for initial setup and environment checks.

```python
%pip install pandas
```

```shell
!ls -la
```

```python
import os
from pathlib import Path
```

--------------------------------

### CLI Commands - Setup

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Commands for installing and setting up the code-review-graph tool.

```APIDOC
## CLI Commands - Setup

### `install`

**Description:** Register MCP server with Claude Code (alias: init).

**Usage:**
```bash
code-review-graph install
```

### `install --dry-run`

**Description:** Preview installation without writing files.

**Usage:**
```bash
code-review-graph install --dry-run
```

```

--------------------------------

### Install and Initialize Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Use this command to install the code-review-graph package and perform initial setup. It is compatible with multiple coding platforms.

```bash
pip install code-review-graph && code-review-graph install
```

--------------------------------

### Configure MCP Server Command

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Example configuration for the .mcp.json file to ensure the server starts correctly using uvx.

```json
"command": "uvx"
"args": ["code-review-graph", "serve"]
```

--------------------------------

### Install and Build code-review-graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Commands to install the package and initialize the codebase graph.

```bash
pip install code-review-graph                     # or: pipx install code-review-graph
code-review-graph install          # auto-detects and configures all supported platforms
code-review-graph build            # parse your codebase
```

--------------------------------

### Install via uv

Source: https://github.com/tirth8205/code-review-graph/blob/main/code-review-graph-vscode/media/walkthrough/install.md

Recommended installation method using the uv package manager.

```bash
uv pip install code-review-graph
```

--------------------------------

### Contributing to Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Setup the development environment by cloning the repository, creating a virtual environment, and installing development dependencies. Run tests to ensure functionality.

```bash
git clone https://github.com/tirth8205/code-review-graph.git
cd code-review-graph
python3 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
pytest
```

--------------------------------

### Clone Repository and Install Dependencies

Source: https://github.com/tirth8205/code-review-graph/blob/main/CONTRIBUTING.md

Clone the repository and install development dependencies using uv. Ensure you have uv installed.

```bash
git clone https://github.com/tirth8205/code-review-graph.git
cd code-review-graph

uv sync --extra dev
```

--------------------------------

### Install via pipx or pip

Source: https://github.com/tirth8205/code-review-graph/blob/main/code-review-graph-vscode/media/walkthrough/install.md

Alternative installation methods using pipx or standard pip.

```bash
pipx install code-review-graph
# or
pip install code-review-graph
```

--------------------------------

### Start MCP Server

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Starts the MCP server using standard input/output for communication.

```bash
code-review-graph serve                        # Start MCP server (stdio)
```

--------------------------------

### Get Docs Section Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/skills/review-pr/SKILL.md

Call this tool before starting to get optimized workflow documentation for the review-pr section. Avoid including full files unless specifically requested.

```python
get_docs_section_tool(section_name="review-pr")
```

--------------------------------

### Install Code Review Graph CLI

Source: https://github.com/tirth8205/code-review-graph/blob/main/code-review-graph-vscode/README.md

Install the Python CLI for the Code Review Graph extension. Recommended to use `uv` for installation. Requires Python 3.10+.

```bash
# Recommended
uv pip install code-review-graph

# Alternatives
pipx install code-review-graph
pip install code-review-graph
```

--------------------------------

### Install and Verify UV

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Commands to install or verify the presence of the uv package manager.

```bash
uv --version
pip install uv
brew install uv
```

--------------------------------

### Install Wiki Generation Dependency

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Install the dependencies required for LLM-powered wiki summaries.

```bash
pip install code-review-graph[wiki]
```

--------------------------------

### Install Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/USAGE.md

Installs the code-review-graph package and configures supported platforms. Run 'code-review-graph install' to auto-detect and configure all platforms, or specify a platform using the --platform flag.

```bash
pip install code-review-graph
code-review-graph install    # auto-detects and configures all supported platforms
code-review-graph build      # parse your codebase
```

```bash
code-review-graph install --platform codex
```

```bash
code-review-graph install --platform cursor
```

```bash
code-review-graph install --platform claude-code
```

--------------------------------

### CLI Commands - Server

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Command for starting the MCP server.

```APIDOC
## CLI Commands - Server

### `serve`

**Description:** Start the MCP server using stdio communication.

**Usage:**
```bash
code-review-graph serve
```

```

--------------------------------

### Get Architecture Overview Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Provides an overview of the project's architecture. `repo_root` can be auto-detected.

```python
get_architecture_overview_tool(
    repo_root: str | None
)
```

--------------------------------

### Install Optional Dependency Groups

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Commands to install various optional feature sets for the project.

```bash
pip install code-review-graph[embeddings]
pip install code-review-graph[google-embeddings]
pip install code-review-graph[communities]
pip install code-review-graph[eval]
pip install code-review-graph[wiki]
pip install code-review-graph[all]
```

--------------------------------

### Verify Setup with Pytest

Source: https://github.com/tirth8205/code-review-graph/blob/main/CONTRIBUTING.md

Run pytest to verify the development setup. This command runs tests quietly and shows a short traceback on failure.

```bash
uv run pytest tests/ --tb=short -q
```

--------------------------------

### Install and Configure Code Review Graph

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Install the package with optional dependencies and configure it for specific AI platforms.

```bash
# Install the package
pip install code-review-graph

# Or with optional dependencies
pip install code-review-graph[embeddings]      # Local vector embeddings
pip install code-review-graph[communities]     # Community detection (igraph)
pip install code-review-graph[all]             # All optional dependencies

# Auto-detect and configure all AI platforms
code-review-graph install

# Or target a specific platform
code-review-graph install --platform cursor
code-review-graph install --platform claude-code
code-review-graph install --platform codex

# Build the knowledge graph
code-review-graph build

# Example output:
# Full build: 500 files, 6285 nodes, 27117 edges (postprocess=full)
```

--------------------------------

### Manage Multi-Repository Setup

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Commands for registering, unregistering, and listing repositories managed by the code-review-graph tool. Aliases can be used for registered repositories.

```bash
code-review-graph register <path> [--alias name]  # Register a repository
```

```bash
code-review-graph unregister <path_or_alias>       # Remove from registry
```

```bash
code-review-graph repos                            # List registered repositories
```

--------------------------------

### Install UV in WSL

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Install the uv package manager specifically within a WSL environment.

```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```

--------------------------------

### Start MCP Server

Source: https://github.com/tirth8205/code-review-graph/blob/main/CLAUDE.md

Launches the FastMCP server, which serves the code graph and registers tools and prompts.

```bash
uv run code-review-graph serve
```

--------------------------------

### Build Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/USAGE.md

Parses your entire codebase to build the initial graph. This is a one-time setup step.

```bash
/code-review-graph:build-graph
```

--------------------------------

### Install Community Detection Dependency

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Install the igraph-based community detection dependency.

```bash
pip install code-review-graph[communities]
```

--------------------------------

### Build Full Knowledge Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/skills/build-graph/SKILL.md

Call this tool with `full_rebuild=True` for first-time setup of the knowledge graph. Ensure the graph status is checked before running.

```python
build_or_update_graph_tool(full_rebuild=True)
```

--------------------------------

### Get Architecture Overview with get_architecture_overview_tool

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Generate an architecture overview based on community structure, highlighting potential coupling issues with recommendations.

```python
result = get_architecture_overview_tool()
```

--------------------------------

### Verify graph installation and status

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Check if the graph is correctly built and registered for the current project session.

```bash
cd /path/to/your/project
code-review-graph status    # should print Files/Nodes/Edges from the built graph
ls .mcp.json                # should exist
cat .mcp.json               # should reference `code-review-graph serve`
# then: fully quit Claude Code and reopen it inside this project
```

--------------------------------

### Install Embeddings for Semantic Search

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/USAGE.md

Installs the necessary package for computing embedding vectors for semantic search. This is required before using embedding-related tools.

```bash
pip install "code-review-graph[embeddings]"
```

--------------------------------

### Install Embeddings Dependency

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Install the necessary dependencies to enable semantic search and vector computation.

```bash
pip install code-review-graph[embeddings]
```

--------------------------------

### Install MCP Server with Claude Code

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Registers the MCP server with Claude Code. Use --dry-run to preview changes without writing files.

```bash
code-review-graph install           # Register MCP server with Claude Code (alias: init)
```

```bash
code-review-graph install --dry-run # Preview without writing files
```

--------------------------------

### Installing Optional Dependencies for Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Install optional dependency groups using pip to enable specific features like embeddings, community detection, or evaluation benchmarks.

```bash
pip install code-review-graph[embeddings]          # Local vector embeddings (sentence-transformers)
```

```bash
pip install code-review-graph[google-embeddings]   # Google Gemini embeddings
```

```bash
pip install code-review-graph[communities]         # Community detection (igraph)
```

```bash
pip install code-review-graph[eval]                # Evaluation benchmarks (matplotlib)
```

```bash
pip install code-review-graph[wiki]                # Wiki generation with LLM summaries (ollama)
```

```bash
pip install code-review-graph[all]                 # All optional dependencies
```

--------------------------------

### CLI Commands for Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Use these commands to manage and interact with the code graph. Commands cover installation, building, updating, and visualizing the graph.

```bash
code-review-graph install          # Auto-detect and configure all platforms
```

```bash
code-review-graph install --platform <name>  # Target a specific platform
```

```bash
code-review-graph build            # Parse entire codebase
```

```bash
code-review-graph update           # Incremental update (changed files only)
```

```bash
code-review-graph status           # Graph statistics
```

```bash
code-review-graph watch            # Auto-update on file changes
```

```bash
code-review-graph visualize        # Generate interactive HTML graph
```

```bash
code-review-graph wiki             # Generate markdown wiki from communities
```

```bash
code-review-graph detect-changes   # Risk-scored change impact analysis
```

```bash
code-review-graph register <path>  # Register repo in multi-repo registry
```

```bash
code-review-graph unregister <id>  # Remove repo from registry
```

```bash
code-review-graph repos            # List registered repositories
```

```bash
code-review-graph eval             # Run evaluation benchmarks
```

```bash
code-review-graph serve            # Start MCP server
```

--------------------------------

### Get Wiki Page Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Retrieves a specific wiki page by `community_name`. `repo_root` can be auto-detected.

```python
get_wiki_page_tool(
    community_name: str          # Community name to look up
    repo_root: str | None
)
```

--------------------------------

### Get Community Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Retrieves details for a specific community. Search by `community_name` or `community_id`. `include_members` determines if community members are listed. `repo_root` can be auto-detected.

```python
get_community_tool(
    community_name: str | None   # Name to search (partial match)
    community_id: int | None     # Database ID
    include_members: bool = False
    repo_root: str | None
)
```

--------------------------------

### Query Graph Tool - Get File Summary

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Obtain a summary of a specific file using the `query_graph_tool`. Set the `pattern` to "file_summary" and provide the `target` file path.

```python
# Get file summary
result = query_graph_tool(pattern="file_summary", target="src/auth/login.py")
```

--------------------------------

### Get Minimal Context Tool

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Retrieve a concise context for a given task, typically used as a first step before employing other graph tools. It provides essential information like risk assessment and suggested tools.

```python
result = get_minimal_context_tool(
    task="review PR #42",
    base="HEAD~1"
)
```

--------------------------------

### List Repositories Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Lists all available repositories. This tool takes no parameters.

```python
list_repos_tool(
    (no parameters)
)
```

--------------------------------

### Build Full Code Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/CLAUDE.md

Initiates a complete build of the code graph from scratch. This is a foundational step before incremental updates.

```bash
uv run code-review-graph build
```

--------------------------------

### Get refactoring suggestions

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Retrieve automated suggestions for code improvements.

```python
result = refactor_tool(mode="suggest")
```

--------------------------------

### Run All Tests with Coverage

Source: https://github.com/tirth8205/code-review-graph/blob/main/CONTRIBUTING.md

Execute all tests and generate a coverage report. Fail the build if coverage is below 50%.

```bash
uv run pytest --cov=code_review_graph --cov-report=term-missing --cov-fail-under=50
```

--------------------------------

### List Registered Repositories

Source: https://github.com/tirth8205/code-review-graph/blob/main/CLAUDE.md

Displays a list of all repositories currently registered in the multi-repo registry.

```bash
uv run code-review-graph repos
```

--------------------------------

### Review changes prompt

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Execute a pre-commit review workflow to identify changes and test gaps.

```python
messages = review_changes(base="HEAD~1")
```

--------------------------------

### Instantiate and Use DataProcessor

Source: https://github.com/tirth8205/code-review-graph/blob/main/tests/fixtures/sample_notebook.ipynb

Creates an instance of the DataProcessor class and calls its process method with an input value. Prints the final output.

```python
processor = DataProcessor('test')
output = processor.process(5)
print(output)
```

--------------------------------

### Configure Specific Platforms

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Commands to target configuration for specific AI coding platforms.

```bash
code-review-graph install --platform codex       # configure only Codex
code-review-graph install --platform cursor      # configure only Cursor
code-review-graph install --platform claude-code  # configure only Claude Code
```

--------------------------------

### Debug Issue Prompt

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Generates a structured prompt for AI assistants to perform guided debugging based on a provided issue description.

```python
# MCP Prompt Call
messages = debug_issue(description="Login timeout after password reset")
# Returns structured prompt for AI assistant to:
# 1. Search for relevant functions
# 2. Trace execution flows
# 3. Check recent changes in affected areas
# 4. Identify potential root causes
```

--------------------------------

### CLI Commands - Wiki

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Command for generating a markdown wiki from code review communities.

```APIDOC
## CLI Commands - Wiki

### `wiki`

**Description:** Generate a markdown wiki from the identified communities.

**Usage:**
```bash
code-review-graph wiki
```

```

--------------------------------

### Register Repository for Multi-repo Search

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/USAGE.md

Registers another repository for cross-repository search. Use the --alias flag to provide a short name for the registered repository.

```bash
code-review-graph register /path/to/other/repo --alias mylib
```

--------------------------------

### Get Documentation Section Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Retrieves a specific section of the documentation. `section_name` can be 'usage', 'review-delta', 'review-pr', 'commands', 'legal', 'watch', 'embeddings', 'languages', or 'troubleshooting'.

```python
get_docs_section_tool(
    section_name: str    # usage, review-delta, review-pr, commands, legal, watch, embeddings, languages, troubleshooting
)
```

--------------------------------

### get_architecture_overview_tool

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Generates a high-level architectural overview including coupling warnings.

```APIDOC
## get_architecture_overview_tool

### Description
Provides an architectural summary, identifying communities and potential coupling issues.

### Response
#### Success Response (200)
- **communities** (array) - List of communities.
- **coupling_warnings** (array) - Warnings regarding high coupling between components.
```

--------------------------------

### Get Flow Details with get_flow_tool

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Retrieve detailed information about a specific execution flow, including the full call path. Can include source code snippets for steps.

```python
result = get_flow_tool(
    flow_id=1,  # Or use flow_name="login" for partial match
    include_source=True
)
```

--------------------------------

### Get Impact Radius Tool

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Analyze the blast radius of changes within the codebase. This tool identifies impacted functions, classes, and files up to a specified `max_depth` and `detail_level`.

```python
result = get_impact_radius_tool(
    changed_files=["src/auth/login.py", "src/utils/validators.py"],
    max_depth=2,
    detail_level="standard"  # or "minimal"
)
```

--------------------------------

### Create Table with SQL

Source: https://github.com/tirth8205/code-review-graph/blob/main/tests/fixtures/sample_databricks_notebook.ipynb

Creates a new table based on a select query from raw data.

```sql
%sql
CREATE TABLE catalog.schema.output AS SELECT * FROM catalog.schema.raw_data
```

--------------------------------

### Get Affected Flows Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Identifies flows affected by code changes. `changed_files` can be auto-detected from git. `base` specifies the Git diff base. `repo_root` can be auto-detected.

```python
get_affected_flows_tool(
    changed_files: list[str] | None  # Auto-detected from git
    base: str = "HEAD~1"
    repo_root: str | None
)
```

--------------------------------

### Get Flow Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Retrieves details for a specific flow. Use `flow_id` (from `list_flows_tool`) or `flow_name` for searching. `include_source` determines if source snippets are included. `repo_root` can be auto-detected.

```python
get_flow_tool(
    flow_id: int | None          # Database ID from list_flows_tool
    flow_name: str | None        # Name to search (partial match)
    include_source: bool = False # Include source snippets for each step
    repo_root: str | None
)
```

--------------------------------

### Environment Variable Configuration

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Configure embedding models, performance tuning, build settings, and data storage paths via environment variables.

```bash
# Embedding configuration
export CRG_EMBEDDING_MODEL="all-MiniLM-L6-v2"  # Default embedding model
export CRG_ACCEPT_CLOUD_EMBEDDINGS=1           # Suppress cloud API warning

# Performance tuning
export CRG_MAX_IMPACT_NODES=500      # Max nodes in impact analysis
export CRG_MAX_IMPACT_DEPTH=3        # Max BFS depth
export CRG_DEPENDENT_HOPS=2          # Multi-hop dependent discovery
export CRG_MAX_SEARCH_RESULTS=100    # Max search results
export CRG_BFS_ENGINE=sql            # "sql" (default) or "networkx"

# Build configuration
export CRG_PARSE_WORKERS=4           # Parallel parsing workers
export CRG_SERIAL_PARSE=1            # Disable parallel parsing
export CRG_GIT_TIMEOUT=30            # Git command timeout (seconds)
export CRG_RECURSE_SUBMODULES=1      # Include git submodules

# Data storage
export CRG_DATA_DIR=/custom/path     # Override default .code-review-graph/
export CRG_REPO_ROOT=/path/to/repo   # Override repo root detection
```

--------------------------------

### Generate Wiki Pages

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/USAGE.md

Creates markdown wiki pages for each detected community within the project. These pages are saved in the .code-review-graph/wiki/ directory.

```bash
code-review-graph wiki
```

--------------------------------

### Embed Graph Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Embeds the knowledge graph using a specified model. Requires `pip install code-review-graph[embeddings]`. `repo_root` can be auto-detected, and `model` specifies the embedding model name.

```python
embed_graph_tool(
    repo_root: str | None
    model: str | None    # Embedding model name
)
```

--------------------------------

### List Graph Statistics Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Lists statistics for the knowledge graph. `repo_root` can be auto-detected.

```python
list_graph_stats_tool(
    repo_root: str | None
)
```

--------------------------------

### Get Impact Radius Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/skills/review-pr/SKILL.md

Analyze the blast radius of changes across the entire pull request. This tool helps identify high-risk areas by highlighting widely depended-upon code.

```python
get_impact_radius_tool(base="main")
```

--------------------------------

### Get Review Context Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/skills/review-pr/SKILL.md

This tool retrieves all changed files across all commits in a pull request, using a specified base branch for diffing. It is essential for obtaining the full context of the review.

```python
get_review_context_tool(base="main")
```

--------------------------------

### Multi-Repository Registry Commands

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Commands to manage multiple repositories for cross-repository operations.

```bash
# Register a repository
code-review-graph register /path/to/project --alias myproject
# Output: Registered: /path/to/project (alias: myproject)

# List all registered repositories
code-review-graph repos
# Output:
#   /path/to/project1  (myproject)
#   /path/to/project2  (api-service)

# Remove from registry
code-review-graph unregister myproject
# Output: Unregistered: myproject
```

--------------------------------

### Get Impact Radius Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Calculates the impact radius of changes. `changed_files` can be auto-detected from git. `max_depth` controls the number of hops in the graph. `repo_root` can be auto-detected, and `base` specifies the Git diff base.

```python
get_impact_radius_tool(
    changed_files: list[str] | None  # Auto-detected from git
    max_depth: int = 2               # Hops in graph
    repo_root: str | None
    base: str = "HEAD~1"
)
```

--------------------------------

### Enable Windows Long Path Support

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/TROUBLESHOOTING.md

Configure Git to support long file paths on native Windows systems.

```bash
git config --system core.longpaths true
```

--------------------------------

### Visualize Code Review Graph

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/USAGE.md

Generates an interactive D3.js force-directed graph of your codebase. The graph starts collapsed and can be expanded by clicking files. Use the search bar to filter and legend to toggle edge visibility.

```bash
code-review-graph visualize
```

```bash
open .code-review-graph/graph.html
```

--------------------------------

### Run Tests for a Single File

Source: https://github.com/tirth8205/code-review-graph/blob/main/CONTRIBUTING.md

Execute tests specifically for the 'test_parser.py' file with verbose output.

```bash
uv run pytest tests/test_parser.py -v
```

--------------------------------

### List Communities Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Lists software communities, sortable by 'size', 'cohesion', or 'name'. `min_size` filters communities by minimum size. `repo_root` can be auto-detected.

```python
list_communities_tool(
    sort_by: str = "size"    # size, cohesion, name
    min_size: int = 0
    repo_root: str | None
)
```

--------------------------------

### Get Review Context Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Retrieves review context, including source code snippets. `changed_files` can be auto-detected from git. `max_depth` controls the graph traversal depth. `include_source` determines if source code is included. `max_lines_per_file` limits the lines per file. `repo_root` can be auto-detected, and `base` specifies the Git diff base.

```python
get_review_context_tool(
    changed_files: list[str] | None
    max_depth: int = 2
    include_source: bool = True
    max_lines_per_file: int = 200
    repo_root: str | None
    base: str = "HEAD~1"
)
```

--------------------------------

### Create a Feature Branch

Source: https://github.com/tirth8205/code-review-graph/blob/main/CONTRIBUTING.md

Create a new Git branch for developing a feature. Replace 'your-feature' with a descriptive name.

```bash
git checkout -b feature/your-feature
```

--------------------------------

### CLI Commands - Multi-repo

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Commands for managing multiple repositories within the code review graph.

```APIDOC
## CLI Commands - Multi-repo

### `register <path> [--alias <name>]`

**Description:** Register a repository with the code review graph. An optional alias can be provided.

**Usage:**
```bash
code-review-graph register /path/to/repo --alias my-repo
```

### `unregister <path_or_alias>`

**Description:** Remove a registered repository from the code review graph.

**Usage:**
```bash
code-review-graph unregister my-repo
```

### `repos`

**Description:** List all currently registered repositories.

**Usage:**
```bash
code-review-graph repos
```

```

--------------------------------

### Register Repository

Source: https://github.com/tirth8205/code-review-graph/blob/main/CLAUDE.md

Adds a repository to the multi-repo registry. This is necessary for managing multiple codebases.

```bash
uv run code-review-graph register <path>
```

--------------------------------

### CLI Commands - Evaluation

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Command for running evaluation benchmarks.

```APIDOC
## CLI Commands - Evaluation

### `eval`

**Description:** Execute evaluation benchmarks for the code review graph.

**Usage:**
```bash
code-review-graph eval
```

```

--------------------------------

### Configuration

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Configuration options for the code-review-graph, including ignore files and optional dependencies.

```APIDOC
## Configuration

### Description
Configuration options for the code-review-graph, including ignore files and optional dependencies.

### Ignore File
To exclude paths from indexing, create a `.code-review-graphignore` file in your repository root.

```
generated/**
*.generated.ts
vendor/**
node_modules/**
```

Note: in git repos, only tracked files are indexed (`git ls-files`), so gitignored files are skipped automatically. Use `.code-review-graphignore` to exclude tracked files or when git isn't available.

### Optional Dependencies

- `pip install code-review-graph[embeddings]`: Local vector embeddings (sentence-transformers)
- `pip install code-review-graph[google-embeddings]`: Google Gemini embeddings
- `pip install code-review-graph[communities]`: Community detection (igraph)
- `pip install code-review-graph[eval]`: Evaluation benchmarks (matplotlib)
- `pip install code-review-graph[wiki]`: Wiki generation with LLM summaries (ollama)
- `pip install code-review-graph[all]`: All optional dependencies
```

--------------------------------

### List Flows Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Lists available flows, sortable by 'criticality', 'depth', 'node_count', 'file_count', or 'name'. `limit` sets the maximum number of results. `kind` can filter by entry point kind (e.g., 'Test', 'Function'). `repo_root` can be auto-detected.

```python
list_flows_tool(
    sort_by: str = "criticality"  # criticality, depth, node_count, file_count, name
    limit: int = 50
    kind: str | None              # Filter by entry point kind (e.g. "Test", "Function")
    repo_root: str | None
)
```

--------------------------------

### Generate Wiki Documentation

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Generates markdown wiki documentation from defined communities within the code review graph.

```bash
code-review-graph wiki                         # Generate markdown wiki from communities
```

--------------------------------

### Generate Wiki Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Generates or regenerates wiki pages for the project. Set `force` to `True` to regenerate all pages regardless of changes. `repo_root` can be auto-detected.

```python
generate_wiki_tool(
    repo_root: str | None
    force: bool = False          # Regenerate all pages even if unchanged
)
```

--------------------------------

### Apply refactoring

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Apply a previously previewed refactoring, optionally using a dry-run mode.

```python
result = apply_refactor_tool(
    refactor_id="abc123",
    dry_run=True  # Preview changes without writing
)
```

```python
result = apply_refactor_tool(refactor_id="abc123", dry_run=False)
```

--------------------------------

### MCP Prompts - Workflow Templates

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Pre-defined workflow templates for common tasks.

```APIDOC
## review_changes

### Description
Pre-commit review workflow using detect_changes, affected_flows, and test gaps.

### Method
PROMPT TEMPLATE

### Endpoint
review_changes

### Parameters
#### Query Parameters
- **base** (str) - Optional - Git diff base (default: HEAD~1)

### Request Example
```json
{
  "base": "main"
}
```

### Response
#### Success Response (200)
(No specific response format defined)

#### Response Example
(No example provided)
```

--------------------------------

### Architecture Map Prompt

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Generates a structured prompt for AI assistants to analyze architecture, communities, and module relationships.

```python
# MCP Prompt Call
messages = architecture_map()
# Returns structured prompt for AI assistant to:
# 1. Run get_architecture_overview_tool
# 2. List top communities and their purposes
# 3. Generate Mermaid diagram of module relationships
# 4. Highlight coupling warnings
```

--------------------------------

### Run Evaluation Benchmarks

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Executes evaluation benchmarks for the code review graph.

```bash
code-review-graph eval                         # Run evaluation benchmarks
```

--------------------------------

### Cross-Repository Search Tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Performs a search across multiple repositories. `query` is the search string. `kind` can filter results (e.g., 'File', 'Class'). `limit` sets the maximum number of results.

```python
cross_repo_search_tool(
    query: str
    kind: str | None
    limit: int = 20
)
```

--------------------------------

### Generate wiki

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Create a markdown wiki based on the current code community structure.

```python
result = generate_wiki_tool(force=False)
```

--------------------------------

### Configuration - Environment Variables

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Environment variables for configuring Code Review Graph behavior.

```APIDOC
## Configuration - Environment Variables

### Description
Set these environment variables to customize the behavior of Code Review Graph.

### Variables
#### Embedding Configuration
- **CRG_EMBEDDING_MODEL** (string) - Default: `all-MiniLM-L6-v2` - Specifies the embedding model to use.
- **CRG_ACCEPT_CLOUD_EMBEDDINGS** (integer) - Default: `0` - Set to `1` to suppress cloud API warnings.

#### Performance Tuning
- **CRG_MAX_IMPACT_NODES** (integer) - Default: `500` - Maximum number of nodes to consider in impact analysis.
- **CRG_MAX_IMPACT_DEPTH** (integer) - Default: `3` - Maximum depth for Breadth-First Search (BFS) in impact analysis.
- **CRG_DEPENDENT_HOPS** (integer) - Default: `2` - Number of hops for multi-hop dependent discovery.
- **CRG_MAX_SEARCH_RESULTS** (integer) - Default: `100` - Maximum number of search results to return.
- **CRG_BFS_ENGINE** (string) - Default: `sql` - The engine to use for BFS. Options: `sql` or `networkx`.

#### Build Configuration
- **CRG_PARSE_WORKERS** (integer) - Default: `4` - Number of parallel workers for parsing code.
- **CRG_SERIAL_PARSE** (integer) - Default: `0` - Set to `1` to disable parallel parsing.
- **CRG_GIT_TIMEOUT** (integer) - Default: `30` - Timeout in seconds for Git commands.
- **CRG_RECURSE_SUBMODULES** (integer) - Default: `0` - Set to `1` to include Git submodules.

#### Data Storage
- **CRG_DATA_DIR** (string) - Default: `.code-review-graph/` - Overrides the default directory for storing graph data.
- **CRG_REPO_ROOT** (string) - Default: Auto-detected - Overrides the repository root detection.

### Example Usage
```bash
export CRG_EMBEDDING_MODEL="all-MiniLM-L6-v2"
export CRG_MAX_IMPACT_NODES=1000
```
```

--------------------------------

### CLI Commands - Build and Update

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/COMMANDS.md

Commands for building and updating the code review graph.

```APIDOC
## CLI Commands - Build and Update

### `build`

**Description:** Perform a full build of the code review graph.

**Usage:**
```bash
code-review-graph build
```

### `update`

**Description:** Perform an incremental update of the code review graph.

**Usage:**
```bash
code-review-graph update
```

### `update --base <ref>`

**Description:** Perform an incremental update with a custom base reference.

**Usage:**
```bash
code-review-graph update --base origin/main
```

```

--------------------------------

### Build and Update Graph CLI Commands

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Commands for performing full builds or incremental updates to the knowledge graph.

```bash
# Full build - parse all files in the repository
code-review-graph build
# Output: Full build: 1122 files, 6285 nodes, 27117 edges (postprocess=full)

# Incremental update - only changed files since last commit
code-review-graph update
# Output: Incremental: 5 files updated, 6290 nodes, 27125 edges (postprocess=full)

# Incremental update with custom base ref
code-review-graph update --base origin/main

# Skip expensive post-processing for faster builds
code-review-graph build --skip-flows          # Skip flow/community detection
code-review-graph build --skip-postprocess    # Skip all post-processing

# Run post-processing separately
code-review-graph postprocess
code-review-graph postprocess --no-flows      # Skip flow detection
code-review-graph postprocess --no-communities # Skip community detection
```

--------------------------------

### Retrieve Review Context

Source: https://github.com/tirth8205/code-review-graph/blob/main/skills/review-delta/SKILL.md

Fetch the necessary context including changed files, impacted nodes, and source snippets.

```python
get_review_context_tool()
```

--------------------------------

### Configuration - Ignore Patterns

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Configure files to exclude from indexing using a `.code-review-graphignore` file.

```APIDOC
## Configuration - Ignore Patterns

### Description
Specify file patterns in a `.code-review-graphignore` file to exclude them from being indexed by Code Review Graph.

### File Format
Uses standard glob patterns, similar to `.gitignore`.

### Example `.code-review-graphignore` Content
```gitignore
generated/**
*.generated.ts
vendor/**
node_modules/**
dist/**
build/**
*.min.js
```
```

--------------------------------

### Ignore Patterns Configuration

Source: https://context7.com/tirth8205/code-review-graph/llms.txt

Define files and directories to exclude from indexing using the .code-review-graphignore file.

```gitignore
# .code-review-graphignore
generated/**
*.generated.ts
vendor/**
node_modules/**
dist/**
build/**
*.min.js
```

--------------------------------

### Contributing

Source: https://github.com/tirth8205/code-review-graph/blob/main/README.md

Instructions for contributing to the code-review-graph project.

```APIDOC
## Contributing

### Setup

```bash
git clone https://github.com/tirth8205/code-review-graph.git
cd code-review-graph
python3 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
pytest
```

### Adding a new language

Edit `code_review_graph/parser.py` and add your extension to `EXTENSION_TO_LANGUAGE` along with node type mappings in `_CLASS_TYPES`, `_FUNCTION_TYPES`, `_IMPORT_TYPES`, and `_CALL_TYPES`. Include a test fixture and open a PR.
```

--------------------------------

### get_minimal_context_tool

Source: https://github.com/tirth8205/code-review-graph/blob/main/docs/LLM-OPTIMIZED-REFERENCE.md

Retrieves a minimal context summary for a given task, including risks, communities, flows, and suggested next tools.

```APIDOC
## POST get_minimal_context_tool

### Description
Returns a concise summary (approx. 100 tokens) to guide the next steps in a code review or analysis task.

### Parameters
#### Request Body
- **task** (string) - Required - The description of the task being performed.
```