LLDB MCP Server

Instructions

Implementation Reference

• lldb_mcp_server.py:815-852 (handler)

  The handler function decorated with @mcp.tool(name="lldb_evaluate"), which evaluates C/C++ expressions using LLDB by creating a target, setting breakpoint, running, executing the expression command, and formatting the output.
  @mcp.tool( name="lldb_evaluate", annotations={ "title": "Evaluate Expression", "readOnlyHint": True, "destructiveHint": False, "idempotentHint": True, "openWorldHint": False, }, ) async def lldb_evaluate(params: EvaluateExpressionInput) -> str: """Evaluate a C/C++ expression in the debugger context. Expressions can include: - Variable access: 'my_var', 'ptr->member' - Array indexing: 'array[5]' - Function calls: 'strlen(str)' - Casts: '(int*)ptr' - Arithmetic: 'x + y * 2' - sizeof: 'sizeof(MyStruct)' Args: params: EvaluateExpressionInput with expression and context Returns: str: Expression result with type information """ commands = [ f"target create {params.executable}", f"breakpoint set --name {params.breakpoint}", "run" + (" " + " ".join(params.args) if params.args else ""), f"expression {params.expression}", "quit", ] result = _run_lldb_script(commands) return f"## Expression: `{params.expression}`\n\n```\n{result['output'].strip()}\n```"
• lldb_mcp_server.py:277-294 (schema)

  Pydantic BaseModel defining the input schema for the lldb_evaluate tool, including executable path, expression to evaluate, breakpoint, and optional args.
  class EvaluateExpressionInput(BaseModel): """Input for evaluating expressions.""" model_config = ConfigDict(str_strip_whitespace=True) executable: str = Field(..., description="Path to the executable", min_length=1) expression: str = Field( ..., description="C/C++ expression to evaluate (e.g., 'sizeof(int)', 'ptr->member', 'array[5]')", min_length=1, ) breakpoint: str = Field( ..., description="Breakpoint location for evaluation context", min_length=1 ) args: list[str] | None = Field( default=None, description="Command-line arguments to pass to the program" )
• lldb_mcp_server.py:118-171 (helper)

  Helper function _run_lldb_script that executes a list of LLDB commands via a temporary script file, capturing output and handling errors. Used by lldb_evaluate to run the LLDB session.
  def _run_lldb_script( commands: list[str], target: str | None = None, working_dir: str | None = None, timeout: int = 60, ) -> dict[str, Any]: """ Execute multiple LLDB commands in sequence. """ cmd = [LLDB_EXECUTABLE] if target: cmd.extend(["--file", target]) cmd.append("--batch") for command in commands: cmd.extend(["-o", command]) try: result = subprocess.run( cmd, capture_output=True, text=True, timeout=timeout, cwd=working_dir or os.getcwd() ) return { "success": result.returncode == 0, "output": result.stdout, "error": result.stderr if result.returncode != 0 else None, "return_code": result.returncode, } except subprocess.TimeoutExpired: return { "success": False, "output": "", "error": f"Commands timed out after {timeout} seconds", "return_code": -1, } except Exception as e: return {"success": False, "output": "", "error": str(e), "return_code": -1} # ============================================================================= # Initialize MCP Server # ============================================================================= mcp = FastMCP(SERVER_NAME) # ============================================================================= # Input Models # ============================================================================= class ResponseFormat(str, Enum): """Output format for tool responses."""