What can you do with this server?

This server enables AI assistants to control STM32 and other embedded MCUs via a J-Link debugger using natural language. Key capabilities include: * Connect/Disconnect: Establish or terminate a connection to a J-Link debugger and target MCU (e.g., STM32F103C8, nRF52840) via SWD or JTAG with configurable speed and retry attempts. * Read RTT Data: Read Real-Time Transfer (RTT) log data from the MCU across one or multiple channels (0–15). * Write RTT Data: Send commands or data to the MCU via RTT channels, supporting plain text and hexadecimal formats. * Set Default RTT Channel: Configure the default RTT channel (0–15) for subsequent operations. * Reset MCU: Reset the target in various modes — normal, auto-reconnect, or halt (pause CPU at reset state). * Flash Firmware: Program .hex or .bin firmware files directly onto the MCU. * Get Device Info: Retrieve detailed information about the connected J-Link debugger and target MCU. * Read/Write Memory: Read or write raw data at arbitrary memory addresses on the MCU (hex or plain string formats). * Start/Stop Log Recording: Begin or stop recording RTT output to a local log file.

How do I use J-Link RTT Viewer MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@J-Link RTT Viewer MCP Connect to my STM32F103C8 board" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

J-Link RTT Viewer MCP

by MisakaMikoto128

Overview Schema Source Related Servers Discussions Admin

Local

J-Link RTT Viewer MCP Server

English | 中文

English

An MCP server that wraps J-Link RTT Viewer, enabling AI assistants to control STM32 development boards via natural language.

Quick Start (3 steps)

# 1. Clone and install
git clone https://github.com/MisakaMikoto128/J-Link-RTT-Viewer-MCP.git
cd J-Link-RTT-Viewer-MCP
python -m venv .venv && .venv\Scripts\activate  # Windows
pip install -r requirements.txt

# 2. Configure Claude Desktop
# Add to %APPDATA%\Claude\claude_desktop_config.json:
# { "mcpServers": { "jlink-rtt": { "command": "python", "args": ["-m", "src.server"], "cwd": "C:\\path\\to\\J-Link-RTT-Viewer-MCP" } } }

# 3. Restart Claude Desktop and use natural language!

Tool Reference

Tool	Parameters	Description
`connect`	`target`, `interface="SWD"`, `speed=4000`	Connect to MCU. Target example: `STM32F103C8`
`disconnect`	(none)	Disconnect from J-Link
`read_rtt`	`channels=[0]`, `max_size=4096`	Read RTT log data from MCU
`write_rtt`	`data`, `channel=0`	Send command to MCU via RTT
`reset`	`mode="normal"`	Reset MCU. Modes: normal, auto_reconnect, halt
`flash_firmware`	`firmware_path`	Flash .hex or .bin file to MCU
`get_device_info`	(none)	Get connected device details
`read_memory`	`address`, `size`	Read MCU memory (hex address like 0x08000000)
`write_memory`	`address`, `data`	Write data to MCU memory
`start_log_recording`	`log_dir="logs"`	Record RTT output to file
`stop_log_recording`	(none)	Stop recording
`set_rtt_channel`	`channel`	Set default RTT channel (0-15)

Common Commands (for write_rtt)

freq 200      - Set LED blink interval to 200ms
freq 1000     - Set LED blink interval to 1s
mode on       - Turn LED on
mode off      - Turn LED off
mode blink    - Set LED to blink mode
status        - Get current settings
ping          - Test connection (returns "pong")
reset         - Reset MCU software

Example: Control LED

User: Connect to my STM32F103C8 board
AI: [connect] Connected to STM32F103C8

User: What's in the RTT logs?
AI: [read_rtt] STM32F103C8T6 LED Control ===\r\nCommands: freq, mode, status...

User: Make the LED blink faster
AI: [write_rtt, data="freq 200"] Done! LED now blinks every 200ms

User: Turn the LED on
AI: [write_rtt, data="mode on"] LED is now on

User: Flash new firmware
AI: [flash_firmware, firmware_path="path/to/firmware.hex"] Flash complete!

Supported Hardware

Debugger: SEGGER J-Link (V7-V10)
MCU: STM32F1xx, STM32F4xx, nRF52xxx, etc.
Interface: SWD, JTAG

Test with MCP Inspector

npx @modelcontextprotocol/inspector -- python -m src.server

Related MCP server: MCP Server for Segger RTT via J-Link

中文

封装 J-Link RTT Viewer 的 MCP 服务器，使 AI 助手能通过自然语言控制 STM32 开发板。

快速开始（3 步）

# 1. 克隆并安装
git clone https://github.com/MisakaMikoto128/J-Link-RTT-Viewer-MCP.git
cd J-Link-RTT-Viewer-MCP
python -m venv .venv && .venv\Scripts\activate  # Windows
pip install -r requirements.txt

# 2. 配置 Claude Desktop
# 添加到 %APPDATA%\Claude\claude_desktop_config.json:
# { "mcpServers": { "jlink-rtt": { "command": "python", "args": ["-m", "src.server"], "cwd": "C:\\path\\to\\J-Link-RTT-Viewer-MCP" } } }

# 3. 重启 Claude Desktop，用自然语言对话！

工具参考

工具	参数	说明
`connect`	`target`, `interface="SWD"`, `speed=4000`	连接 MCU。示例: `STM32F103C8`
`disconnect`	(无)	断开 J-Link
`read_rtt`	`channels=[0]`, `max_size=4096`	读取 MCU 的 RTT 日志
`write_rtt`	`data`, `channel=0`	通过 RTT 发送命令到 MCU
`reset`	`mode="normal"`	重置 MCU。模式: normal, auto_reconnect, halt
`flash_firmware`	`firmware_path`	烧录 .hex 或 .bin 文件
`get_device_info`	(无)	获取设备详细信息
`read_memory`	`address`, `size`	读取 MCU 内存
`write_memory`	`address`, `data`	写入 MCU 内存
`start_log_recording`	`log_dir="logs"`	录制 RTT 输出到文件
`stop_log_recording`	(无)	停止录制
`set_rtt_channel`	`channel`	设置默认 RTT 通道 (0-15)

常用命令（用于 write_rtt）

freq 200      - 设置 LED 闪烁间隔为 200ms
freq 1000     - 设置 LED 闪烁间隔为 1s
mode on       - 打开 LED
mode off      - 关闭 LED
mode blink    - 设置 LED 闪烁模式
status        - 获取当前设置
ping          - 测试连接（返回 "pong"）
reset         - 软件重置 MCU

示例：控制 LED

用户: 连接到我的 STM32F103C8 开发板
AI: [connect] 已连接到 STM32F103C8

用户: RTT 日志里有什么？
AI: [read_rtt] STM32F103C8T6 LED Control ===\r\nCommands: freq, mode, status...

用户: 让 LED 闪快一点
AI: [write_rtt, data="freq 200"] 完成！LED 现在每 200ms 闪烁一次

用户: 打开 LED
AI: [write_rtt, data="mode on"] LED 已打开

用户: 烧录新固件
AI: [flash_firmware, firmware_path="path/to/firmware.hex"] 烧录完成！

支持的硬件

调试器: SEGGER J-Link (V7-V10)
MCU: STM32F1xx, STM32F4xx, nRF52xxx 等
接口: SWD, JTAG

使用 MCP Inspector 测试

npx @modelcontextprotocol/inspector -- python -m src.server

Available Tools

12 tools

connectA

连接到 J-Link 调试器和目标 MCU

Args: target: 目标设备型号 (如 STM32F103C8, STM32F407VG, nRF52840) interface: 调试接口 (SWD 或 JTAG)，默认 SWD speed: 连接速度 (kHz)，默认 4000 channel: RTT 通道号，默认 0 retry_attempts: 重试次数，默认 3

Returns: 连接状态和设备信息的 JSON 字符串

Inputs: target (required), interface, speed, channel, +1 more. Returns: result

disconnectA

断开 J-Link 连接

安全地断开与 J-Link 调试器的连接，清理所有资源。

Returns: 断开状态的 JSON 字符串

Inputs: no named parameters. Returns: result

flash_firmwareA

烧录固件到 MCU

将固件文件烧录到连接的目标 MCU。

Args: firmware_path: 固件文件路径 (.hex, .bin)

Returns: 烧录结果的 JSON 字符串

Inputs: firmware_path (required). Returns: result

get_device_infoA

获取当前连接设备的信息

返回 J-Link 调试器和目标 MCU 的详细信息。

Returns: 设备信息的 JSON 字符串

Inputs: no named parameters. Returns: result

read_memoryA

读取 MCU 内存

从目标 MCU 的指定地址读取内存数据。

Args: address: 起始地址 (如 0x08000000) size: 读取字节数 (1-1048576)

Returns: 内存数据的 JSON 字符串（十六进制格式）

Inputs: address (required), size (required). Returns: result

read_rttB

读取 RTT 日志数据

从目标 MCU 读取 RTT (Real-Time Transfer) 日志数据。默认读取所有通道 (0-15)，返回结构化的日志数据。

Args: channels: 要读取的通道列表，None 表示所有通道 (默认) max_size: 最大读取字节数，默认 4096

Returns: RTT 数据的 JSON 字符串，包含时间戳、通道、内容和元数据

Inputs: channels, max_size. Returns: result

resetA

重置目标 MCU

重置连接的目标 MCU。支持多种重置模式。

Args: mode: 重置模式 - "normal": 标准重置，保留连接会话 (默认) - "auto_reconnect": 重置后自动断开重连 - "halt": 重置后暂停 CPU（停在复位状态） wait_time: 重置后等待时间（秒），默认 0.3

Returns: 重置结果的 JSON 字符串

Inputs: mode, wait_time. Returns: result

set_rtt_channelB

设置默认 RTT 通道

设置后续 RTT 操作使用的默认通道。

Args: channel: 通道号 (0-15)

Returns: 设置结果的 JSON 字符串

Inputs: channel (required). Returns: result

start_log_recordingC

开始日志记录

开始将 RTT 数据记录到文件。

Args: log_dir: 日志目录，默认 "logs"

Returns: 记录状态的 JSON 字符串

Inputs: log_dir. Returns: result

stop_log_recordingB

停止日志记录

停止将 RTT 数据记录到文件。

Returns: 停止状态的 JSON 字符串

Inputs: no named parameters. Returns: result

write_memoryB

写入 MCU 内存

向目标 MCU 的指定地址写入数据。

Args: address: 起始地址 (如 0x08000000) data: 要写入的数据（十六进制字符串或普通字符串） is_hex: 是否为十六进制数据，默认 True

Returns: 写入结果的 JSON 字符串

Inputs: address (required), data (required), is_hex. Returns: result

write_rttA

向 RTT 通道写入数据

向目标 MCU 的 RTT 通道发送数据。

Args: data: 要写入的数据 channel: 通道号，None 使用默认通道 is_hex: 是否为十六进制数据，默认 False

Returns: 写入结果的 JSON 字符串

Inputs: data (required), channel, is_hex. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

TDQS

A3.7/5.0

Scored 2026-06-21 06:45 across 12 tools

Disambiguation5/5

Each tool targets a distinct operation (connect, disconnect, flash, memory read/write, RTT read/write, logging, reset, device info) with no overlapping purposes.

Naming Consistency5/5

All tools use consistent verb_noun snake_case naming (e.g., read_memory, write_rtt) with no deviations or mixed conventions.

Tool Count5/5

12 tools is well-scoped for a J-Link/RTT debugger, covering essential operations without being excessive or minimal.

Completeness4/5

Core operations (connect, flash, memory R/W, RTT R/W, logging, reset) are covered, but minor gaps like erase or verify exist.

Maintenance

ActivityInactive

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

DNS-DATA RemoteDeprecated
com.dns-data
Connect AI assistants to authorized Windows and macOS devices for files, commands, and management.
File Systems Command Line OS Automation
11
https://mcp.dns-data.com/mcp
GitHub
io.github.github
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
Version Control Developer Tools CI/CD & DevOps
1234,365
https://api.githubcopilot.com/mcp/
ZionDesk
app.ziondesk
Connect ChatGPT, Claude, Cursor or VS Code to your Windows PC: files, terminal, browser, apps.
File Systems Command Line Browser Automation
1
https://mcp.ziondesk.app/mcp
Remote Desktop CommanderA
app.desktopcommander
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Shell Access OS Automation
322220
https://mcp.desktopcommander.app/mcp

Related MCP Servers

jlink-mcp
Embedded system Developer Tools
xun123456
A
license
A
quality
D
maintenance
Enables AI assistants like Claude to directly debug microcontrollers via JLink, supporting breakpoints, single-step, memory/register access, variable inspection, RTT logging, and firmware flashing.
Updated 6 months ago (2026-05-25 05:37 UTC)
25
7
MIT
MCP Server for Segger RTT via
Embedded system Developer Tools
ForeverJin
A
license
A
quality
D
maintenance
Enables LLMs to interact with embedded devices by reading and writing Segger RTT data through a J-Link debugger.
Updated a month ago (2026-10-08 06:52 UTC)
9
1
MIT
jlink-mcp
Embedded system Developer Tools
Klievan
A
license
B
quality
B
maintenance
Enables AI assistants to debug embedded devices via SEGGER J-Link probes, including reading memory, flashing firmware, streaming RTT logs, and diagnosing crashes.
Updated 17 days ago (2026-10-26 11:28 UTC)
48
181 npm
43
MIT
stmctl-mcp
Embedded system Command Line
Sathvik-smarttrak
A
license
A
quality
D
maintenance
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.
Updated 3 months ago (2026-07-30 13:32 UTC)
12
MIT

J-Link RTT Viewer MCP Server

English

Quick Start (3 steps)

Tool Reference

Common Commands (for write_rtt)

Example: Control LED

Supported Hardware

Test with MCP Inspector

中文

快速开始（3 步）

工具参考

常用命令（用于 write_rtt）

示例：控制 LED

支持的硬件

使用 MCP Inspector 测试

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

jlink-mcp

MCP Server for Segger RTT via

jlink-mcp

stmctl-mcp