获取当前调用堆栈
AI agents call get_stack_trace to retrieve information from WinDbg GUI MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves the current call stack from the WinDbg debugger session. Call stack inspection is a fundamental read-only debugging operation that does not modify state, execute code, or cause destructive changes. It simply queries and returns stack frame information for analysis purposes. The blast radius of misuse is minimal—an AI agent cannot cause damage by reading call stack data.
From the tool's definition Tool name 'get_stack_trace' and description '获取当前调用堆栈' (Get current call stack) indicate retrieval of debugging information with no side effects. Call stack inspection is a read-only operation in debuggers.
Attacks that exploit this kind of access
获取当前调用堆栈. It is categorised as a Read tool in the WinDbg GUI MCP Server MCP Server, which means it retrieves data without modifying state.
Register the WinDbg GUI MCP Server MCP server in PolicyLayer and add a rule for get_stack_trace: 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 WinDbg GUI MCP Server. Nothing to install.
get_stack_trace is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_stack_trace 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.
Set action: deny in the PolicyLayer policy for get_stack_trace. 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.
get_stack_trace is provided by the WinDbg GUI MCP Server MCP server (jianqiaojia/windbg-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →