// SCAN REPORT

Rr

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

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

Last updated: 30 Jun 2026

31 can modify or destroy data

17 read-only

48 tools total

Community server · catalogue entry verified 30/06/2026

How to control Rr ↓

TOOL MAP

What Rr exposes to your agents

Read (17) Write / Execute (29) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Rr tools

Destructive · High breakpoint_delete This is a destructive operation because it permanently deletes debugging state. Destructive · High checkpoint_delete Deleting checkpoints removes saved execution states that cannot be recovered. Execute · High finish The finish command triggers execution of the target program to a specific point (function return), which is a debugger control operation that causes the program to run. Execute · Critical gdb_raw GDB (GNU Debugger) commands can inspect, modify, and execute code within a debugged process.

31 of Rr's 48 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Rr

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

Deny destructive operations

{
  "breakpoint_delete": {
    "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

{
  "breakpoint_disable": {
    "limits": [
      {
        "counter": "breakpoint_disable_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "args": {
    "limits": [
      {
        "counter": "args_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 Rr — 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 RR →

Instant setup, no code required.

FULL CATALOGUE

All 48 Rr tools

DESTRUCTIVE 2 tools

Destructive breakpoint_delete Delete a breakpoint permanently. Get breakpoint_id Destructive checkpoint_delete Delete a checkpoint to free resources. Each checkpoint consumes memory

EXECUTE 24 tools

Execute breakpoint_set Set a breakpoint to pause execution at a specific Execute checkpoint_restore Restore a previously created checkpoint, instantly Execute continue Continue execution forward until hitting a breakpoint, signal, or program end. Execute finish Run until the current function returns, then stop at the caller. Execute frame_select Switch to a different stack frame to inspect its Execute gdb_raw Execute an arbitrary GDB command (escape hatch). Execute handle_signal Configure how GDB handles a specific signal (stop, pass to program, print). Execute info Run a GDB Execute interrupt Interrupt a running program to pause execution. Execute next Step forward one source line, stepping over function Execute nexti Step forward one machine instruction, stepping over call instructions. Execute print Evaluate any expression in the current context and return its value. Execute reverse_continue Continue execution backward (rr Execute reverse_finish Run backward to the point where the current function was called (function entry). Execute reverse_next Step backward one source line, stepping over function calls in reverse. Execute reverse_nexti Step backward one machine instruction, stepping over calls in reverse. Execute reverse_step Step backward one source line, entering into function calls in reverse. Execute reverse_stepi Step backward one machine instruction. Reverse counterpart of stepi. Execute run_to_event Run to a specific event number (rr global time). Execute session_create Create a new replay session for debugging a specific Execute step Step forward one source line, entering into function calls. Execute stepi Step forward one machine instruction. Use for Execute thread_select Switch to a different thread for inspection. Execute watchpoint_set Set a hardware watchpoint to break when a variable

WRITE 5 tools

Write breakpoint_disable Temporarily disable a breakpoint without deleting Write breakpoint_enable Re-enable a disabled breakpoint. Useful when you Write catch Set a catchpoint to break on C++ exceptions or syscalls. Write checkpoint_create Create a checkpoint (snapshot) at the current position. Checkpoints let you Write session_close End a replay session and clean up its GDB/rr process.

READ 17 tools

Read args Get function arguments for the current frame. Read backtrace Get the call stack showing how execution reached Read breakpoint_list List all breakpoints and watchpoints in the session Read checkpoint_list List all checkpoints in a session with their IDs Read examine_memory Examine raw memory at an address. Use to inspect Read find_in_memory Search memory for a byte pattern between two addresses. Read locals Get all local variables in the current stack frame with their values and types. Read registers Get CPU register values (rax, rbx, rsp, rip, etc.). Read session_list List all active replay sessions with their IDs, traces, and PIDs. Read source_files List all source files known to the debugger. Read source_list List source code around a location. Use to see Read source_path Get the current source file path and line number. Read threads_list List all threads in the current process with their IDs and current stack frames. Read trace_info Get detailed metadata about a specific trace Read trace_processes List all processes in a trace with PID, parent PID, command, and exit status. Read traces_list List all available rr recordings on the system. Read when Get current execution position as an rr event

EXPLORE

Related servers

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

FAQ

Questions about Rr

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

Yes. The Rr server exposes 2 destructive tools including breakpoint_delete, checkpoint_delete. 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 Rr? +

The Rr server has 5 write tools including breakpoint_disable, breakpoint_enable, catch. 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 Rr.

How many tools does the Rr MCP server expose? +

48 tools across 4 categories: Destructive, Execute, Read, Write. 17 are read-only. 31 can modify, create, or delete data.

How do I enforce a policy on Rr? +

Register the Rr 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 Rr tool call.

Deterministic rules across all 48 Rr tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON RR →

Instant setup, no code required.

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