将任务分配给另一个 Agent。对方收到 task_assigned 事件后会自主开始执行,无需人工确认。
AI agents invoke assign_task to trigger actions in Agent-Comm-Hub. 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.
This tool triggers autonomous execution by another agent without any human-in-the-loop confirmation. The downstream effects depend entirely on what task is assigned, and since execution is automatic and immediate, misuse could cascade across multiple agents in a multi-agent system, warranting high severity.
From the tool's definition "对方收到 task_assigned 事件后会自主开始执行,无需人工确认" (the receiving agent autonomously begins execution upon receiving the task_assigned event, with no human confirmation required)
Attacks that exploit this kind of access
将任务分配给另一个 Agent。对方收到 task_assigned 事件后会自主开始执行,无需人工确认。. It is categorised as a Execute tool in the Agent-Comm-Hub MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Agent-Comm-Hub MCP server in PolicyLayer and add a rule for assign_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 Agent-Comm-Hub. Nothing to install.
assign_task is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the assign_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.
Set action: deny in the PolicyLayer policy for assign_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.
assign_task is provided by the Agent-Comm-Hub MCP server (liuboacean/agent-comm-hub). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →