https://github.com/muzain187/tg_mcp
🔗 A lightweight Python interface that exposes TigerGraph operations (queries, schema, vertices, edges, UDFs) as structured tools and URI-based resources for MCP agents.
https://github.com/muzain187/tg_mcp
Last synced: 3 months ago
JSON representation
🔗 A lightweight Python interface that exposes TigerGraph operations (queries, schema, vertices, edges, UDFs) as structured tools and URI-based resources for MCP agents.
- Host: GitHub
- URL: https://github.com/muzain187/tg_mcp
- Owner: Muzain187
- License: mit
- Created: 2025-04-30T05:41:45.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2025-05-03T05:43:15.000Z (5 months ago)
- Last Synced: 2025-06-17T13:16:08.886Z (4 months ago)
- Language: Python
- Size: 26.4 KB
- Stars: 5
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# TG_MCP
A lightweight Python interface that exposes TigerGraph operations (queries, schema, vertices, edges, UDFs) as structured tools and URI-based resources for MCP agents.
## Table of Contents
1. [Features](#features)
2. [Project Structure](#project-structure)
3. [Installation](#installation)
4. [Configuration](#configuration)
5. [Connecting to Claude](#connecting-to-claude)
6. [Examples](#examples)
7. [Contributing](#contributing)
8. [License](#license)
## Features
- **Schema Introspection**
Retrieve full graph schema (vertex & edge types).
- **Query Execution**
Run installed GSQL queries or raw GSQL strings with parameters.
- **Vertex & Edge Upsert**
Create or update vertices and edges programmatically.
- **Resource URIs**
Access graph objects through `tgraph://vertex/...` and `tgraph://query/...` URIs.
- **UDF & Algorithm Listing**
Fetch installed user-defined functions and GDS algorithm catalogs.
## Project Structure
```
TG_MCP/
├── config.py # Environment config (HOST, GRAPH, SECRET)
├── tg_client.py # Encapsulates TigerGraphConnection and core operations
├── tg_tools.py # `@mcp.tool` definitions exposing client methods
├── tg_resources.py # `@mcp.resource` URI handlers
├── main.py # MCP app bootstrap (`mcp.run()`)
├── pyproject.toml # Project metadata & dependencies
├── LICENSE # MIT License
└── .gitignore # OS/Python ignore rules
```
## Installation
1. **Clone the repo**
```bash
git clone https://github.com/Muzain187/TG_MCP.git
cd TG_MCP
```
2. **Create & activate a virtual environment**
```bash
python3 -m venv venv
source venv/bin/activate
```
3. **Install dependencies**
```bash
pip install .
```
> Requires `mcp[cli]>=1.6.0` and `pyTigerGraph>=1.8.6`.
## Configuration
Set the following environment variables before running:
```bash
export TG_HOST=https://
export TG_GRAPH=
export TG_SECRET=
```
These are read by `config.py`.
## Connecting to Claude
This MCP server can be installed into the **Claude Desktop** client so that Claude can invoke your TigerGraph tools directly:
```bash
uv run mcp install main.py
```
After running the above, restart Claude Desktop and you’ll see your MCP tools available via the hammer 🛠icon.
## Examples:
## Contributing
1. Fork the repository
2. Create a feature branch
```bash
git checkout -b feature/YourFeature
```
3. Commit your changes
```bash
git commit -m "Add YourFeature"
```
4. Push to branch
```bash
git push origin feature/YourFeature
```
5. Open a Pull Request
Please ensure all new code is covered by tests and follows PEP-8 style.
## License
This project is licensed under the **MIT License**.