SimosMCP
MCP server for Simos 18.1 ECU tuning that reads, writes, diffs, and validates XDF/BIN files.
README
SimosMCP
MCP server for Simos 18.1 ECU tuning — read, write, diff, and validate XDF/BIN files.
What it does
Reads XDF definition files and BIN binary files, letting you:
- List all tables and scalars in an XDF
- Read table data (2D maps) with proper axis labels and scaled values
- Read scalar values (single parameters)
- Write table data back to BIN files
- Write scalar values back to BIN files
- Diff two BIN files to see what changed
- Validate tables against their min/max ranges
- Search for tables by name
Setup
No dependencies beyond Python 3.10+. Uses only standard library (xml.etree.ElementTree, struct).
VS Code MCP config
Add to .vscode/mcp.json:
{
"servers": {
"SimosMCP": {
"type": "stdio",
"command": "python",
"args": ["path/to/simosmcp/server.py"]
}
}
}
Or run directly
cd path/to/simosmcp
python server.py
Tools
| Tool | Description |
|---|---|
list_tables |
List all tables/scalars, filter by category or search |
read_table |
Read a table's data from a BIN (returns x/y axes + 2D data) |
read_scalar |
Read a scalar/constant value |
write_table |
Write a 2D array to a table in a BIN |
write_scalar |
Write a value to a scalar in a BIN |
diff_bins |
Compare two BINs and show differences |
validate_table |
Check a table's values against min/max ranges |
search_tables |
Search for tables/scalars by name |
How it works
The XDF file is an XML definition that tells the parser:
- Where data lives in the BIN (addresses, data sizes)
- How to interpret it (signed/unsigned, endianness)
- How to scale raw values to human-readable units (MATH equations)
The parser handles:
- Embedded axis data (read from BIN)
- Label-based axes (defined in XDF)
- Linked axes (cross-references between tables)
- MATH equation transformation (forward and inverse)
- Both ECU (Simos 18.1) and TCU (DQ250) XDF formats
- Automatic endianness detection from
mmedtypeflags
Tested with
SC8S50_switchpatch29.33_v1.006.xdf(ECU, Simos 18.1)F45M_DSG_v1.007.xdf(TCU, DQ250)
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。