AI agents invoke eval_to_send_to_process to trigger actions in Pwndbg. 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.
The name 'eval_to_send_to_process' indicates arbitrary evaluation of input that is then sent to a running process. In the context of a debugger MCP server for ELF binaries, this almost certainly allows executing arbitrary expressions or commands within or directed at a running process.
From the tool's definition Tool name 'eval_to_send_to_process' strongly implies evaluating/executing something and sending it to a process; server context is 'debug ELF binaries' with tools like 'interrupt_process' and 'debug_control'
Documented attack patterns abuse exactly the kind of access eval_to_send_to_process gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pwndbg, and nothing reaches the server without passing your rules. This is the rule we recommend for eval_to_send_to_process:
{
"version": "1",
"default": "deny",
"tools": {
"eval_to_send_to_process": {
"limits": [
{
"counter": "eval_to_send_to_process_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}eval_to_send_to_process stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
eval_to_send_to_process. It is categorised as a Execute tool in the Pwndbg MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Pwndbg MCP server in PolicyLayer and add a rule for eval_to_send_to_process: 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. Nothing to install.
eval_to_send_to_process 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 eval_to_send_to_process 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 eval_to_send_to_process. 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.
eval_to_send_to_process is provided by the Pwndbg MCP server (rocketmadev/pwndbg-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 17 Pwndbg tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
17 Pwndbg tools catalogued and risk-classified — across an index of 42,500+ MCP servers.