artifact_safety_report · Model Safety MCP: Risk & Policy

WHAT IT DOES

What artifact_safety_report does on Model Safety MCP

AI agents invoke artifact_safety_report to trigger actions in Model Safety MCP. 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 artifact_safety_report needs a policy

This tool executes automated safety scanning workflows on machine learning artifacts. While not destructive or modifying data, it runs analysis code/processes on user-provided models, making it Execute rather than Read. The severity is medium because scanning itself is benign, but misuse could analyze malicious models or trigger resource-intensive operations.

From the tool's definition Tool performs 'Run the broadest available model artifact safety report' which involves executing a scanning operation against model artifacts. This is a runtime analysis/inspection action that triggers external scanning logic rather than simple data retrieval.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

artifact_safety_report Rate-limited

artifact_safety_report 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": {
    "artifact_safety_report": {
      "limits": [
        {
          "counter": "artifact_safety_report_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 Model Safety MCP, so you can run agents against it safely.

FAQ

Questions about artifact_safety_report

What does the artifact_safety_report tool do? +

Run the broadest available model artifact safety report. It is categorised as a Execute tool in the Model Safety MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on artifact_safety_report? +

Register the Model Safety MCP server in PolicyLayer and add a rule for artifact_safety_report: 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 Model Safety MCP. Nothing to install.

What risk level is artifact_safety_report? +

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

Can I rate-limit artifact_safety_report? +

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

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

artifact_safety_report is provided by the Model Safety MCP server (itsalissonsilva/modelsafetymcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.