execute-task · Back-Agent MCP Server MCP: Risk & Policy

WHAT IT DOES

What execute-task does on Back-Agent MCP Server

AI agents invoke execute-task to trigger actions in Back-Agent MCP Server. 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 execute-task needs a policy

This tool runs code and external operations (code generation, task completion) whose effects depend on the task arguments provided. While the disclaimer emphasizes it is 'NOT a direct shell command executor,' it clearly executes development tasks in working directories via an external CLI. This fits Execute rather than Write because the effects are operational triggers rather than simple data creation/modification.

From the tool's definition Tool executes development tasks through Claude Code CLI in specified working directories. Description states it enables 'automated code generation and task completion' by 'bridging Claude Desktop with Claude Code CLI.' This is a task execution mechanism that…

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

execute-task Rate-limited

execute-task 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": {
    "execute-task": {
      "limits": [
        {
          "counter": "execute-task_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 Back-Agent MCP Server, so you can run agents against it safely.

FAQ

Questions about execute-task

What does the execute-task tool do? +

IMPORTANT: This is an AI programming assistant tool, NOT a direct shell command executor.\n\n. It is categorised as a Execute tool in the Back-Agent MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on execute-task? +

Register the Back-Agent MCP Server MCP server in PolicyLayer and add a rule for execute-task: 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 Back-Agent MCP Server. Nothing to install.

What risk level is execute-task? +

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

Can I rate-limit execute-task? +

Yes. Add a rate_limit block to the execute-task 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 execute-task completely? +

Set action: deny in the PolicyLayer policy for execute-task. 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 execute-task? +

execute-task is provided by the Back-Agent MCP Server MCP server (zuens2020/back-agent-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.