stmctl-mcp
Enables AI assistants to flash firmware, program memory, modify option bytes, erase chips, reset boards, and capture SWO printf traces for STM32 microcontrollers via STM32CubeCLT.
README
STM32 Flashing & Hardware Control MCP Server (stmctl-mcp)
A production-grade Model Context Protocol (MCP) server wrapping STM32CubeCLT (STM32_Programmer_CLI.exe). This MCP server enables any AI coding assistant or agent framework (Antigravity, Claude Desktop, Cursor, custom Node.js/Python agents) to perform hardware flashing, memory dumps, option byte modification, chip erase, board resets, and live SWO printf log tracing.
Hardware & Software Requirements
1. Host Computer Requirements
- Node.js: v18.0.0 or higher
- Operating System: Windows 10 / 11, Linux, or macOS
2. Platform Installation Scripts (scripts/)
Automated installer & environment setup scripts are provided in the scripts/ folder:
- Windows:
scripts/install_win.bat - Linux:
scripts/install_linux.sh(Includes ST-LINK USBudevpermission rules setup) - macOS:
scripts/install_mac.sh(Includes Homebrewstlinkfallback check)
Flexible Path Configuration & Dynamic Resolution Hierarchy
To accommodate different computers, OS environments, and version numbers (e.g., STM32CubeCLT_1.17.0, 1.22.0, 2.0.0), stmctl-mcp resolves STM32_Programmer_CLI.exe using the following priority sequence:
Priority 1: Local Configuration File (stmctl_config.json)
You can manually specify any custom path in stmctl_config.json in the server root directory:
{
"cli_path": "C:\\MyCustomPath\\STM32_Programmer_CLI.exe",
"default_port": "SWD",
"default_mode": "HOTPLUG"
}
Priority 2: Environment Variable
Set STM32_PROGRAMMER_CLI_PATH in your environment or MCP launcher configuration:
STM32_PROGRAMMER_CLI_PATH=C:\ST\STM32CubeCLT_1.22.0\STM32CubeProgrammer\bin\STM32_Programmer_CLI.exe
Priority 3: Automatic Dynamic Version Scanning (C:\ST\STM32CubeCLT_*)
If no manual path is specified, stmctl-mcp automatically scans C:\ST\ for any installed version number (sorting highest version first) to locate:
C:\ST\STM32CubeCLT_<version>\STM32CubeProgrammer\bin\STM32_Programmer_CLI.exe
Priority 4: Standard Program Files Directories
Scans standard Program Files, Program Files (x86), /usr/local/STMicroelectronics/..., /opt/st/..., and /Applications/... directories.
Priority 5: System PATH Fallback
Tries executing STM32_Programmer_CLI directly from system environment PATH.
Capabilities & Tools Exposed
stm32_list_probes: Discover connected ST-LINK, J-Link, UART COM ports, USB devices.stm32_connect_info: Query chip ID, Flash/RAM size, revision ID, CPU core.stm32_flash_firmware: Program.bin,.hex,.elf,.srecfiles with automatic erase, verification, reset, and run flags.stm32_erase_memory: Perform full chip mass erase or sector/page erase.stm32_read_memory: Read raw memory addresses or dump to.bin/.hexfiles.stm32_write_memory: Write bytes/words to specific memory addresses or peripheral registers.stm32_read_option_bytes: Read RDP protection level, BOR level, Watchdog, and Boot flags.stm32_write_option_bytes: Modify Option Bytes (e.g.RDP=0xAA,nBOOT0=1).stm32_reset_mcu: Hardware (HWrst), Software (SWrst), or Core (Crst) reset.stm32_run_mcu: Start or resume program execution at specified address.stm32_start_swv_trace: Stream/capture live SWO Serial Wire Viewer printf logs.stm32_raw_cli: Direct passthrough for any customSTM32_Programmer_CLIflags.
Installation & Build
1. Install Dependencies
npm install
2. Build TypeScript Source
npm run build
This compiles TypeScript files into ./dist/index.js.
How to Integrate stmctl-mcp into Any Agent
Method 1: Desktop IDEs & Apps (Antigravity, Claude Desktop, Cursor, VS Code)
Add the JSON configuration block below to your application's MCP settings file (e.g., claude_desktop_config.json or .vscode/mcp.json):
{
"mcpServers": {
"stmctl-mcp": {
"command": "node",
"args": [
"c:/Users/kunur/OneDrive/Documents/Smarttrak/Firmware_Agent/STMCTL_MCP/dist/index.js"
]
}
}
}
Method 2: Custom Node.js / TypeScript Agent
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js";
async function main() {
const transport = new StdioClientTransport({
command: "node",
args: ["c:/Users/kunur/OneDrive/Documents/Smarttrak/Firmware_Agent/STMCTL_MCP/dist/index.js"],
});
const client = new Client({ name: "firmware-agent-host", version: "1.0.0" }, { capabilities: {} });
await client.connect(transport);
const probeResult = await client.callTool({
name: "stm32_list_probes",
arguments: {},
});
console.log("Probes Result:", probeResult.content[0].text);
}
main().catch(console.error);
Method 3: Custom Python Agent (LangChain / LlamaIndex / CrewAI / AutoGen)
import asyncio
from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
async def run_firmware_agent():
server_params = StdioServerParameters(
command="node",
args=["c:/Users/kunur/OneDrive/Documents/Smarttrak/Firmware_Agent/STMCTL_MCP/dist/index.js"],
)
async with stdio_client(server_params) as (read, write):
async with ClientSession(read, write) as session:
await session.initialize()
result = await session.call_tool("stm32_list_probes", arguments={})
print("Probe Result:", result.content[0].text)
asyncio.run(run_firmware_agent())
Method 4: Interactive Web Browser Debugging (MCP Inspector)
npm run inspector
This opens the MCP Inspector UI at http://localhost:5173.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.