Set a breakpoint at an offset relative to the base of main ($main).
AI agents invoke switch_break_at to trigger actions in Gdb Multiarch. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
Setting a breakpoint modifies the state of the debugger/target process by inserting a trap instruction or registering a halt condition. This is an active operation that affects execution flow of the target program, making it Execute. It is reversible (breakpoints can be deleted), so not Destructive. Severity is medium because misuse can pause or disrupt the target process but doesn't directly destroy data.
From the tool's definition Set a breakpoint at an offset relative to the base of main ($main)
Attacks that exploit this kind of access
Set a breakpoint at an offset relative to the base of main ($main). It is categorised as a Execute tool in the Gdb Multiarch MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Gdb Multiarch MCP server in PolicyLayer and add a rule for switch_break_at: 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 Multiarch. Nothing to install.
switch_break_at 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 switch_break_at 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 switch_break_at. 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.
switch_break_at is provided by the Gdb Multiarch MCP server (sbergeron42/gdb-multiarch-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 →