// SCAN REPORT

MCP Debug Server

43 tools. 30 can modify or destroy data without limits.

2 destructive tools with no built-in limits. Policy required.

Last updated: 29 Jun 2026

30 can modify or destroy data

13 read-only

43 tools total

Community server · catalogue entry verified 29/06/2026

How to control MCP Debug Server ↓

TOOL MAP

What MCP Debug Server exposes to your agents

Read (13) Write / Execute (28) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous MCP Debug Server tools

Destructive · Medium remove_breakpoint Removing a breakpoint is an irreversible action within the debugging session; once removed, the breakpoint configuration is lost and execution will no longer pause at that location. Destructive · Medium remove_hardware_breakpoint Removing a hardware breakpoint is an irreversible action that permanently deletes a debugging control point. Execute · Critical execute_command This tool permits execution of arbitrary WinDbg commands without apparent filtering or restriction. Execute · High start_debug_server This tool triggers an external operation (launching CDB in server mode on a specified TCP port) whose effects depend on the port argument and optional process attachment.

30 of MCP Debug Server's 43 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control MCP Debug Server

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

Deny destructive operations

{
  "remove_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

{
  "disable_breakpoint": {
    "limits": [
      {
        "counter": "disable_breakpoint_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "display_symbol": {
    "limits": [
      {
        "counter": "display_symbol_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 MCP Debug 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 MCP DEBUG →

Instant setup, no code required.

FULL CATALOGUE

All 43 MCP Debug Server tools

DESTRUCTIVE 2 tools

Destructive remove_breakpoint Remove breakpoint by ID Destructive remove_hardware_breakpoint Remove a hardware breakpoint

EXECUTE 24 tools

Execute attach_process Attach debugger to a process by PID or name (optional if already attached during server start) Execute break_execution Break into debugger (pause process execution) Execute connect_debugger Connect CDB client to running CDB server session Execute create_minidump Generate minidump on demand Execute detach_process Detach debugger from current process Execute disconnect_debugger Disconnect debugger client and close session Execute enable_breakpoint Enable a disabled breakpoint Execute enable_exception_monitoring Enable background exception monitoring with auto-capture Execute evaluate_expression Evaluate expression or resolve symbol address Execute execute_command Execute any raw WinDbg command directly Execute reload_symbols Reload symbols for all modules Execute resume_execution Resume process execution (go/continue) Execute resume_with_monitoring Resume execution with exception monitoring Execute set_breakpoint Set breakpoint at address or symbol Execute set_exception_break Set exception breakpoint with auto-capture Execute start_debug_server Start CDB in server mode on specified TCP port, optionally attached to a process Execute step_into Step into next instruction (trace) Execute step_out Step out of current function Execute step_over Step over next instruction Execute stop_debug_server Stop debug server running on specified port Execute switch_thread Switch debugger context to specific thread Execute toggle_hardware_breakpoint Enable or disable a hardware breakpoint without removing it Execute write_memory Write data to memory at specified address Execute write_register Write value to a CPU register

WRITE 4 tools

Write disable_breakpoint Disable a breakpoint without removing it Write disassemble Disassemble code at address Write set_hardware_breakpoint Set hardware breakpoint using CPU debug registers (fast, non-intrusive, works with fast threads) Write set_symbol_path Set symbol search path

READ 13 tools

Read display_symbol Display symbol information (resolve wildcards) Read get_callstack Get current call stack (stack trace) Read get_exception_info Get last exception/event information Read get_last_exception_captured Get the last captured exception data (even after thread termination) Read get_process_info Get current process information Read list_breakpoints List all active breakpoints Read list_hardware_breakpoints List all active hardware breakpoints with hit counts and status Read list_modules List all loaded modules/DLLs Read list_threads List all threads in target process Read module_info Get detailed information about a specific module Read read_memory Read memory at specified address Read read_registers Read CPU register values Read search_memory Search memory for a pattern

EXPLORE

Related servers

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

FAQ

Questions about MCP Debug Server

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

Yes. The MCP Debug Server server exposes 2 destructive tools including remove_breakpoint, remove_hardware_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 MCP Debug Server? +

The MCP Debug Server server has 4 write tools including disable_breakpoint, disassemble, set_hardware_breakpoint. 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 MCP Debug Server.

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

43 tools across 4 categories: Destructive, Execute, Read, Write. 13 are read-only. 30 can modify, create, or delete data.

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

Register the MCP Debug 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 MCP Debug Server tool call.

Deterministic rules across all 43 MCP Debug Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MCP DEBUG →

Instant setup, no code required.

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