// SCAN REPORT

GDB MCP Server

31 tools. 19 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 3 Jul 2026

19 can modify or destroy data

12 read-only

31 tools total

Community server · catalogue entry verified 03/07/2026

How to control GDB MCP Server ↓

TOOL MAP

What GDB MCP Server exposes to your agents

Read (12) Write / Execute (18) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous GDB MCP Server tools

Destructive · Low gdb_delete_breakpoint Deleting a breakpoint is an irreversible removal of a debugging artifact. Execute · Critical gdb_command This tool enables execution of arbitrary GDB MI commands, which can trigger external operations with wide-ranging effects depending on arguments: running/stopping target processes, modifying memory, setting breakpoints, Execute · High gdb_run gdb_run executes arbitrary code on a debugged embedded system. Execute · Medium gdb_connect This tool establishes a TCP connection to a remote gdbserver or QEMU instance, triggering an external network operation.

19 of GDB MCP Server's 31 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control GDB MCP Server

PolicyLayer is an MCP gateway — it sits between your AI agents and GDB MCP Server, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "gdb_delete_breakpoint": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "gdb_disable_breakpoint": {
    "limits": [
      {
        "counter": "gdb_disable_breakpoint_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "gdb_disassemble": {
    "limits": [
      {
        "counter": "gdb_disassemble_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register GDB MCP Server — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON GDB →

Instant setup, no code required.

FULL CATALOGUE

All 31 GDB MCP Server tools

DESTRUCTIVE 1 tools

Destructive gdb_delete_breakpoint 删除指定编号的断点。

EXECUTE 15 tools

Execute gdb_command 直接执行GDB Machine Interface (MI)命令。参数: - command: MI命令 (以 - 开头，如 -break-list, -exec-run) - waitForDone (可选): 是 Execute gdb_connect 通过TCP连接到目标板的gdbserver或QEMU。参数: - target: 目标地址，格式为 host:port (例如 192.168.1.100:1234 或 localhost:1234) 示例: - Execute gdb_continue 从当前位置继续执行程序。程序将运行直到遇到断点或异常。 Execute gdb_disconnect 优雅地断开与目标板的连接并终止GDB进程。会先发送detach命令通知gdbserver。 Execute gdb_enable_breakpoint 启用指定编号的断点。 Execute gdb_evaluate_expression 计算并返回表达式的值。参数: - expression: 要计算的表达式 (如 x, ptr, sizeof(var), x+y) 示例: - gdb_evaluate_expression({expression Execute gdb_finish 继续执行直到当前函数返回。在返回点暂停。 Execute gdb_init 启动指定架构的GDB进程，默认使用aarch64交叉编译器。参数: - gdbPath (可选): GDB可执行文件路径 - architecture (可选): 目标架构 (aarch64/arm/x86_64) Execute gdb_interrupt 中断正在运行的程序。程序将暂停在当前位置。 Execute gdb_load_program 加载要调试的可执行文件及其符号表。参数: - programPath: 可执行文件路径必须在连接目标板后调用此命令。 Execute gdb_next 执行一行代码。如果当前行包含函数调用，将跳过该函数（不进入内部）。 Execute gdb_run 从程序入口点开始执行。程序将运行直到遇到断点或异常。 Execute gdb_select_thread 切换到指定线程进行调试。 Execute gdb_step 执行一行代码。如果当前行包含函数调用，将进入该函数内部。 Execute gdb_write_register 修改指定寄存器的值。参数: - register: 寄存器名称 (如 r0, pc, sp) - value: 要写入的值 (如 0x1000 或 42) 警告: 此操作可能影响程序执行状态。

WRITE 3 tools

Write gdb_disable_breakpoint 禁用指定编号的断点。 Write gdb_set_breakpoint 在指定位置设置断点，支持条件断点和临时断点。参数: - location: 断点位置 (文件:行号或函数名) - condition (可选): 条件表达式，如 x>5, i==10 - temporary (可 Write gdb_set_condition 为已有断点设置或修改条件表达式。参数: - breakpointId: 断点编号 - condition: 条件表达式，如 x>5, i==10；传空字符串清空条件示例: - gdb_set_condition(

READ 12 tools

Read gdb_disassemble 反汇编当前函数或指定地址范围。参数: - start (可选): 起始地址 - end (可选): 结束地址默认反汇编当前函数。 Read gdb_get_current_location 获取程序当前位置（源文件和行号）。 Read gdb_list_arguments 列出当前函数的参数及其值。 Read gdb_list_breakpoints 列出当前设置的所有断点及其详细信息（包括条件）。 Read gdb_list_frames 列出当前的调用栈帧信息。 Read gdb_list_locals 列出当前栈帧的局部变量及其值。 Read gdb_list_registers 获取目标架构的所有寄存器名称列表。 Read gdb_list_source_files 列出调试程序的所有源文件。 Read gdb_list_threads 列出当前程序的所有线程信息。 Read gdb_read_memory 从指定地址读取内存内容。参数: - address: 内存地址 (如 0x400000) - length: 读取字节数 (1-4096) Read gdb_read_registers 读取指定寄存器的值，默认读取所有寄存器。参数: - registers (可选): 寄存器名称列表，如 [r0, pc, sp] Read gdb_select_frame 选择指定编号的栈帧，用于查看该层的局部变量。参数: - frameId: 栈帧编号 (从0开始，0为当前帧)

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about GDB MCP Server

Can an AI agent delete data through the GDB MCP Server MCP server? +

Yes. The GDB MCP Server server exposes 1 destructive tools including gdb_delete_breakpoint. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through GDB MCP Server? +

The GDB MCP Server server has 3 write tools including gdb_disable_breakpoint, gdb_set_breakpoint, gdb_set_condition. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach GDB MCP Server.

How many tools does the GDB MCP Server MCP server expose? +

31 tools across 4 categories: Destructive, Execute, Read, Write. 12 are read-only. 19 can modify, create, or delete data.

How do I enforce a policy on GDB MCP Server? +

Register the GDB MCP Server MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every GDB MCP Server tool call.

Deterministic rules across all 31 GDB MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON GDB →

Instant setup, no code required.

31 GDB MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.