llm_complete · VibeServe MCP: Risk & Policy

WHAT IT DOES

What llm_complete does on VibeServe

AI agents invoke llm_complete to trigger actions in VibeServe. 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.

WHY IT NEEDS A POLICY

Why llm_complete needs a policy

This tool executes an external LLM API call with arbitrary prompt input. The effects depend entirely on the prompt arguments, which could include generating and returning malicious code, sensitive data exfiltration prompts, or driving downstream pipeline steps in unintended ways.

From the tool's definition Complete a prompt through Big-Homie — triggers an LLM completion call to an external service

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

llm_complete Rate-limited

llm_complete 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.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "llm_complete": {
      "limits": [
        {
          "counter": "llm_complete_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches VibeServe, so you can run agents against it safely.

FAQ

Questions about llm_complete

What does the llm_complete tool do? +

Complete a prompt through Big-Homie. It is categorised as a Execute tool in the VibeServe MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on llm_complete? +

Register the VibeServe MCP server in PolicyLayer and add a rule for llm_complete: 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 VibeServe. Nothing to install.

What risk level is llm_complete? +

llm_complete is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit llm_complete? +

Yes. Add a rate_limit block to the llm_complete 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.

How do I block llm_complete completely? +

Set action: deny in the PolicyLayer policy for llm_complete. 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.

What MCP server provides llm_complete? +

llm_complete is provided by the VibeServe MCP server (ncsound919/vibeserve). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.