stop_agent · Containerized Strands Agents MCP: Risk & Policy

WHAT IT DOES

What stop_agent does on Containerized Strands Agents

AI agents invoke stop_agent to trigger actions in Containerized Strands Agents. 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 stop_agent needs a policy

Stopping a Docker container is an Execute action: it triggers an external operation (container termination) whose effects are contingent on arguments (which agent). While container shutdown is technically reversible (restart possible), the immediate termination itself qualifies as Execute rather than Write because it is a control action over an external system with potentially disruptive side effects on agent…

From the tool's definition 'Stop an agent's Docker container immediately' — this terminates a running process with immediate effect, which is an external operation whose consequences depend on which agent is targeted and its current state.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

stop_agent Rate-limited

stop_agent 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": {
    "stop_agent": {
      "limits": [
        {
          "counter": "stop_agent_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 Containerized Strands Agents, so you can run agents against it safely.

FAQ

Questions about stop_agent

What does the stop_agent tool do? +

Stop an agent's Docker container immediately. It is categorised as a Execute tool in the Containerized Strands Agents MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on stop_agent? +

Register the Containerized Strands Agents MCP server in PolicyLayer and add a rule for stop_agent: 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 Containerized Strands Agents. Nothing to install.

What risk level is stop_agent? +

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

Can I rate-limit stop_agent? +

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

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

stop_agent is provided by the Containerized Strands Agents MCP server (mkmeral/containerized-strands-agents). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.