Check for incoming messages. Call this periodically to see if other agents have sent you requests.
AI agents call check_inbox to retrieve information from Interagent without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves message state without modifying, deleting, or executing any operations. It is purely informational, allowing an agent to poll for incoming communications. The action is read-only with no blast radius beyond exposing message contents that are presumably already intended for the requesting agent.
From the tool's definition Tool description states 'Check for incoming messages' and 'see if other agents have sent you requests' - this is a retrieval operation with no side effects. The verb 'check' and 'see' indicate passive information gathering.
Attacks that exploit this kind of access
Check for incoming messages. Call this periodically to see if other agents have sent you requests. It is categorised as a Read tool in the Interagent MCP Server, which means it retrieves data without modifying state.
Register the Interagent MCP server in PolicyLayer and add a rule for check_inbox: 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 Interagent. Nothing to install.
check_inbox is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the check_inbox 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 check_inbox. 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.
check_inbox is provided by the Interagent MCP server (signalclaude/interagent). 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 →