AI agents invoke ask_question to trigger actions in Mcp Llm. 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 invokes an external LLM service, constituting an external operation with effects (API calls, compute costs, potential for prompt injection or data exfiltration). It is not a simple read from a local store — it executes a request against an external system.
From the tool's definition "Ask a question to the LLM" — triggers an external LLM operation whose output depends on the question argument
Documented attack patterns abuse exactly the kind of access ask_question gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mcp Llm, and nothing reaches the server without passing your rules. This is the rule we recommend for ask_question:
{
"version": "1",
"default": "deny",
"tools": {
"ask_question": {
"limits": [
{
"counter": "ask_question_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}ask_question 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.
Ask a question to the LLM. It is categorised as a Execute tool in the Mcp Llm MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mcp Llm MCP server in PolicyLayer and add a rule for ask_question: 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 Mcp Llm. Nothing to install.
ask_question 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 ask_question 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 ask_question. 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.
ask_question is provided by the Mcp Llm MCP server (sammcj/mcp-llm). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 4 Mcp Llm tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4 Mcp Llm tools catalogued and risk-classified — across an index of 42,500+ MCP servers.