debug_attach

THE RISK

High Risk

Why debug_attach needs a policy

This tool attaches a debugger to a running embedded target process. It triggers an external operation (GDB/OpenOCD debug session attachment) without modifying or deleting data, but it does interact with and gain control over a running system, which constitutes executing an external operation. Misuse could allow inspection or manipulation of a live embedded system.

From the tool's definition Attach to a running target without resetting or downloading firmware. Uses launch.json config with request=

Documented attack patterns abuse exactly the kind of access debug_attach gives an agent:

POLICY

How to control debug_attach

PolicyLayer is an MCP gateway — it sits between your AI agents and Openocd, and nothing reaches the server without passing your rules. This is the rule we recommend for debug_attach:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "debug_attach": {
      "limits": [
        {
          "counter": "debug_attach_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

debug_attach stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Openocd — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

RATE-LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Openocd tools

Destructive shutdown Gracefully shut down the MCP server. Stops debug session if active, then exits. Execute debug_command Execute one GDB command in current active debug session. Execute debug_start Start debug session using specified launch configuration. Execute debug_stop Stop current active debug session and terminate OpenOCD/GDB processes. Execute debug_continue Continue target execution (asynchronous — returns immediately). GDB stays responsive after Execute debug_interrupt Interrupt/pause the running target via GDB/MI -exec-interrupt with platform-specific fallb Execute flash_download Flash firmware once using specified launch configuration without starting debug session. Write set_project Set current project directory and load debug configurations from .vscode/launch.json.

All 14 Openocd tools →

Execute tools on other servers

Android Forensics ADB MCP Server adb_shell_command EMBA-MCP run_emba_scan Frida Game Hacking MCP remove_breakpoint Procmon request_elevation

Go deeper

High-risk tools in Openocd → Every high-risk tool this server exposes, in one view.
All high-risk MCP tools → Every high-risk tool across the catalogue, and the policy pattern that contains them.
The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Agent sandbox → Bounding what an agent can execute and where the boundary should sit.
Least privilege for MCP → Granting agents the minimum tool surface the task needs.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about debug_attach

What does the debug_attach tool do? +

Attach to a running target without resetting or downloading firmware. Uses launch.json config with request=. It is categorised as a Execute tool in the Openocd MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on debug_attach? +

Register the Openocd MCP server in PolicyLayer and add a rule for debug_attach: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Openocd. Nothing to install.

What risk level is debug_attach? +

debug_attach is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit debug_attach? +

Yes. Add a rate_limit block to the debug_attach rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block debug_attach completely? +

Set action: deny in the PolicyLayer policy for debug_attach. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides debug_attach? +

debug_attach is provided by the Openocd MCP server (luiox/openocd-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Openocd tool call.

Start from Openocd, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →