infer · Parallelix MCP: Risk & Policy

WHAT IT DOES

What infer does on Parallelix

AI agents invoke infer to trigger actions in Parallelix. 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 infer needs a policy

This tool triggers external execution of inference operations on a remote compute network. While the output is read-only (completion text), the act of running arbitrary prompts constitutes code execution with unpredictable side effects depending on prompt content (e.g., prompts could request harmful outputs, data extraction, or other malicious tasks).

From the tool's definition Tool 'infer' runs a prompt on a distributed network ('Run a single prompt on the ParalleliX network'), which executes code/logic whose effects depend entirely on the prompt arguments provided by the user.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

infer Rate-limited

infer 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": {
    "infer": {
      "limits": [
        {
          "counter": "infer_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 Parallelix, so you can run agents against it safely.

FAQ

Questions about infer

What does the infer tool do? +

Run a single prompt on the ParalleliX network. Returns the completion plus the serving node id and Proof-of-Execution hash. For bulk independent work, prefer parallel_map. It is categorised as a Execute tool in the Parallelix MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on infer? +

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

What risk level is infer? +

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

Can I rate-limit infer? +

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

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

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