Ask the connected LLM a question using sampling
AI agents invoke ask_llm to trigger actions in MCP TypeScript Starter. 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 triggers an external operation — invoking an LLM via sampling — whose effects depend entirely on the arguments passed. It is not a simple read (it initiates an active model call), and it could be used to execute arbitrary prompts, extract information, or chain further actions through the LLM. This places it in the Execute category.
From the tool's definition Ask the connected LLM a question using sampling
Documented attack patterns abuse exactly the kind of access ask_llm gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCP TypeScript Starter, and nothing reaches the server without passing your rules. This is the rule we recommend for ask_llm:
{
"version": "1",
"default": "deny",
"tools": {
"ask_llm": {
"limits": [
{
"counter": "ask_llm_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}ask_llm 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 the connected LLM a question using sampling. It is categorised as a Execute tool in the MCP TypeScript Starter MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MCP TypeScript Starter MCP server in PolicyLayer and add a rule for ask_llm: 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 TypeScript Starter. Nothing to install.
ask_llm 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_llm 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_llm. 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_llm is provided by the MCP TypeScript Starter MCP server (sammorrowdrums/mcp-typescript-starter). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MCP TypeScript Starter, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
8 MCP TypeScript Starter tools catalogued and risk-classified — across an index of 43,000+ MCP servers.