# Ansys Lumerical Mcp

> MCP server for Ansys Lumerical automation via ansys-lumerical-core.

- **Type:** MCP server
- **Install:** `agentstack add mcp-sheepwangzz-ansys-lumerical-mcp`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [SheepWangZz](https://agentstack.voostack.com/s/sheepwangzz)
- **Installs:** 0
- **Category:** [Integrations](https://agentstack.voostack.com/c/integrations)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [SheepWangZz](https://github.com/SheepWangZz)
- **Source:** https://github.com/SheepWangZz/ansys-lumerical-mcp

## Install

```sh
agentstack add mcp-sheepwangzz-ansys-lumerical-mcp
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Ansys Lumerical MCP Server

MCP server for automating Ansys Lumerical through `ansys-lumerical-core`.

## Installation

Requires Python `>=3.10,=0.2.0`
- `mcp>=1.27.0`

With conda:

```powershell
conda create -n ansys-lumerical-mcp python=3.13 pip
conda activate ansys-lumerical-mcp
python -m pip install ansys-lumerical-mcp
```

From a source checkout, conda users can also create an environment with:

```powershell
conda env create -f environment.yml
conda activate ansys-lumerical-mcp
```

That source environment installs this project in editable mode and lets pip
resolve the runtime dependencies.

## Development

```powershell
python -m pip install -e ".[tests]"
python -m ansys_lumerical_mcp
```

The server uses the MCP SDK's stdio transport. The first implemented tool is
`lumerical_status`, which checks the Python package and discovered local
Lumerical installation without launching a solver session.

Implemented tools:

- `lumerical_status`
- `list_available_products`
- `open_session`
- `list_sessions`
- `close_session`
- `run_script`
- `put_variable`
- `get_variable`

## Project layout

```text
ansys-lumerical-mcp/
├─ docs/
├─ examples/
├─ src/
│  └─ ansys_lumerical_mcp/
├─ tests/
├─ pyproject.toml
├─ README.md
├─ LICENSE
└─ .gitignore
```

## Build

```powershell
python -m pip install -e ".[build]"
python -m build
```

## MCP client configuration

The same configuration is also available at
`examples/client_config.json`. See `docs/mcp_client_configuration.md` for the
full setup notes, including Codex configuration and verification commands.

```json
{
  "mcpServers": {
    "ansys-lumerical": {
      "command": "ansys-lumerical-mcp",
      "env": {
        "LUMERICAL_MCP_INSTALL_DIR": "C:\\Path\\To\\Lumerical\\vXXX"
      }
    }
  }
}
```

`LUMERICAL_MCP_INSTALL_DIR` is optional when `ansys-lumerical-core`
auto-discovers the local installation. If discovery works, remove the `env`
block entirely. Keep machine-specific paths in your local MCP client settings,
not in committed project files.

For Codex, the installed-package configuration is:

```toml
[mcp_servers.ansys-lumerical]
command = "ansys-lumerical-mcp"
```

## Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [SheepWangZz](https://github.com/SheepWangZz)
- **Source:** [SheepWangZz/ansys-lumerical-mcp](https://github.com/SheepWangZz/ansys-lumerical-mcp)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/mcp-sheepwangzz-ansys-lumerical-mcp
- Seller: https://agentstack.voostack.com/s/sheepwangzz
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.