pwndbg_canary
AI agents call pwndbg_canary to retrieve information from Pwndbg Lldb without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
In pwndbg, the 'canary' command reads and displays the stack canary value from the current process being debugged. This is a read/query operation with no side effects. However, since the description is empty, confidence is reduced. The sibling tools (backtrace, bins, auxv, aslr) are all read-only inspection commands, supporting this classification.
From the tool's definition Tool name 'pwndbg_canary' with empty description; based on pwndbg context, 'canary' typically reads the stack canary value from a debugged process
Attacks that exploit this kind of access
pwndbg_canary. It is categorised as a Read tool in the Pwndbg Lldb MCP Server, which means it retrieves data without modifying state.
Register the Pwndbg Lldb MCP server in PolicyLayer and add a rule for pwndbg_canary: 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 Pwndbg Lldb. Nothing to install.
pwndbg_canary 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 pwndbg_canary 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 pwndbg_canary. 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.
pwndbg_canary is provided by the Pwndbg Lldb MCP server (micro-evaluation-group/pwndbg-lldb-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 →