relay_run · RelayPlane MCP: Risk & Policy

WHAT IT DOES

What relay_run does on RelayPlane

AI agents invoke relay_run to trigger actions in RelayPlane. 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 relay_run needs a policy

This tool invokes third-party AI models with user-supplied prompts, which constitutes executing external operations. The output and side effects (token consumption, provider charges) are determined by argument content.

From the tool's definition Tool description states it 'Execute[s] a single AI model call' and 'Returns output, token usage, estimated provider cost, and trace URL.' The verb 'Execute' and the ability to trigger external LLM provider operations (OpenAI/Anthropic) demonstrate…

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

relay_run Rate-limited

relay_run 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": {
    "relay_run": {
      "limits": [
        {
          "counter": "relay_run_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 RelayPlane, so you can run agents against it safely.

FAQ

Questions about relay_run

What does the relay_run tool do? +

Execute a single AI model call. Useful for testing prompts before building full workflows. Returns output, token usage, estimated provider cost, and trace URL. Note: Cost tracks your provider bill (OpenAI/Anthropic), not RelayPlane fees - we. It is categorised as a Execute tool in the RelayPlane MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on relay_run? +

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

What risk level is relay_run? +

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

Can I rate-limit relay_run? +

Yes. Add a rate_limit block to the relay_run 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 relay_run completely? +

Set action: deny in the PolicyLayer policy for relay_run. 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 relay_run? +

relay_run is provided by the RelayPlane MCP server (relayplane/mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.