执行一行代码。如果当前行包含函数调用,将进入该函数内部。
AI agents invoke gdb_step to trigger actions in GDB MCP Server. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes program code in the debugger (GDB step instruction), which runs arbitrary code on the target system. It triggers execution whose effects depend on what code is being stepped through, fitting the Execute category. Severity is medium because while it only executes one step at a time, repeated misuse could drive the program into unintended states or trigger side effects on the embedded target.
From the tool's definition 执行一行代码。如果当前行包含函数调用,将进入该函数内部。 (Execute one line of code; if the current line contains a function call, step into the function)
Attacks that exploit this kind of access
执行一行代码。如果当前行包含函数调用,将进入该函数内部。. It is categorised as a Execute tool in the GDB MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the GDB MCP Server MCP server in PolicyLayer and add a rule for gdb_step: 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 GDB MCP Server. Nothing to install.
gdb_step is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the gdb_step 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 gdb_step. 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.
gdb_step is provided by the GDB MCP Server MCP server (zjm1060/gdb-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 →