CONTAM MCP
A Windows-based MCP server that exposes NIST's CONTAM simulation tools and ContamX bridge-mode controls to AI agents. It enables automated airflow and contaminant transport modeling, including project inspection, simulation execution, and real-time bridge session management.
README
CONTAM MCP
CONTAM MCP is a Windows-first MCP server that exposes CONTAM command-line tools and ContamX bridge-mode controls to AI agents.
The project is designed for public use. It packages stable CONTAM automation workflows behind a standard MCP tool surface so an agent can inspect projects, diagnose broken inputs, run simulations, compare outputs, and drive bridge sessions over time.
The design is informed by the CONTAM API direction described in the Building Simulation paper "Development and Application of CONTAM APIs" and by the official ContamX documentation and bridge protocol materials from NIST.
Why This Project Exists
CONTAM is powerful, but most automation workflows still start from desktop usage patterns or manual case handling. This repository turns the practical parts of the CONTAM toolchain into an MCP server so agents can:
- inspect
.prjmodels before running them - catch missing weather or support-file references
- run
contamx3.exeand collect outputs - compare
.simfiles - export
simreadtext - control ContamX bridge-mode sessions step by step
Host Compatibility
This server is not Codex-only.
Any host that can launch a local stdio MCP server should be able to use it. Confirmed target hosts documented in this repository include:
- Codex Desktop / Codex Windows App
- Claude Code
- Claude Desktop
- Cursor
- other MCP-capable local hosts
Host-specific setup examples are in:
Easiest Connection Option
For public users, the easiest setup is now an npx-launched local server sourced from this repository.
Use:
command: npx
args: -y --package github:summer521521/contam_MCP contam-mcp
This avoids manually pointing a host at a local server.js path.
Main Capabilities
- discover bundled CONTAM executables
- find
.prj,.sim,.wth,.ctm, and related files - inspect project metadata and external file references
- diagnose broken project references
- update
.prjfile references - run
contamx3.exe - upgrade old
.prjfiles withprjup.exe - compare
.simfiles withsimcomp.exe - export
simreadtext output - start, inspect, advance, and close ContamX bridge sessions
- adjust zones, junctions, ambient targets, AHS settings, and control nodes by ID or by name
Five-Minute Quickstart
- Configure your MCP host to run
npx -y --package github:summer521521/contam_MCP contam-mcp. - Restart the host.
- Ask the host to inspect or run a sample
.prj.
If you prefer a local clone, that path is still supported and documented in the host guide.
For the full step-by-step tutorial, see:
Example Prompts
Call discover_contam_installation and confirm CONTAM is available.List CONTAM case files in this folder.Inspect this PRJ file and summarize its references and date range.Run a test input only check for this PRJ.Run this PRJ and list the generated outputs.Start a CONTAM bridge session for this project.List the zones in the active bridge session.Advance the active bridge session by 300 seconds and return path flow updates.Close the active bridge session.
Repository Layout
contam-mcp/: MCP server source, developer guide, and regression scriptsdocs/: public quickstart and host setup guides.github/workflows/: GitHub Actions workflows- repository root: bundled CONTAM executables and supporting DLLs
Privacy and CI
This repository includes a privacy check that scans tracked files for personal filesystem paths before public sharing.
Run it locally with:
cd contam-mcp
npm run privacy:check
GitHub Actions also runs this check automatically.
References
- NIST CONTAM Download Page
- NIST CONTAM Documentation Page
- Dols, Shen, Polidoro, et al. "Development and Application of CONTAM APIs" (Building Simulation, 2026)
- CONTAM User Guide and Program Documentation Version 3.4 (NIST TN 1887r1)
- Host Setup Guide
For Maintainers
If you want to extend the server, review bridge protocol coverage, or run the official regression suite, start here:
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。