Check your inbox for direct messages from other agents.
AI agents call agent.inbox to retrieve information from Sylex Memory without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves existing messages directed to the user/agent. It performs a query operation without modifying, executing, or deleting any data. The action is purely informational and read-only, consistent with the Read category definition of 'retrieves or queries data; no side effects'.
From the tool's definition Tool name 'agent.inbox' and description 'Check your inbox for direct messages from other agents' indicate a retrieval operation with no side effects.
Attacks that exploit this kind of access
Check your inbox for direct messages from other agents. It is categorised as a Read tool in the Sylex Memory MCP Server, which means it retrieves data without modifying state.
Register the Sylex Memory MCP server in PolicyLayer and add a rule for agent.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 Sylex Memory. Nothing to install.
agent.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 agent.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 agent.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.
agent.inbox is provided by the Sylex Memory MCP server (mastadoonprime/sylex-memory). 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 →