close_case · Decision OS MCP: Risk & Policy

WHAT IT DOES

What close_case does on Decision OS MCP

AI agents use close_case to create or update resources in Decision OS MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Decision OS MCP environment.

WHY IT NEEDS A POLICY

Why close_case needs a policy

This tool modifies data (changes case status from open to closed) reversibly—cases can be reopened or corrected. It does not delete data or create financial obligations, and while it triggers a workflow (logging outcomes), it does not execute arbitrary code or system commands.

From the tool's definition close_case closes a case with outcome signals and regret scoring; description explicitly states 'Call this when work is complete,' indicating it modifies the state of a case record by marking it as closed.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

close_case Rate-limited

close_case stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "close_case": {
      "limits": [
        {
          "counter": "close_case_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

LIMIT THIS TOOL →

Instant setup, no code required.

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

FAQ

Questions about close_case

What does the close_case tool do? +

Close a case with outcome signals. Call this when work is complete. The regret score (0-3) is critical: - 0: Would choose the same approach again - 1: Minor improvements possible - 2: Significant regret, different approach likely better - 3: Strong regret, wrong posture/approach. It is categorised as a Write tool in the Decision OS MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on close_case? +

Register the Decision OS MCP server in PolicyLayer and add a rule for close_case: 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 Decision OS MCP. Nothing to install.

What risk level is close_case? +

close_case is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit close_case? +

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

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

close_case is provided by the Decision OS MCP server (marianstefi20/decision-os-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.