Long-poll: blocks up to
AI agents call wait_for_message to retrieve information from Mcp Switchboard without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Long-polling is a read/receive pattern: the tool waits for a message to arrive and returns it. No data is created, modified, or deleted. The description is truncated, slightly lowering confidence, but the 'long-poll' and 'blocks up to' phrasing clearly indicates a blocking read operation. Severity is low as misuse would at worst cause the agent to hang waiting for a message.
From the tool's definition 'Long-poll: blocks up to' — tool is described as a long-polling operation, which waits/blocks to retrieve incoming messages without modifying state
Attacks that exploit this kind of access
Long-poll: blocks up to. It is categorised as a Read tool in the Mcp Switchboard MCP Server, which means it retrieves data without modifying state.
Register the Mcp Switchboard MCP server in PolicyLayer and add a rule for wait_for_message: 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 Mcp Switchboard. Nothing to install.
wait_for_message 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 wait_for_message 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 wait_for_message. 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.
wait_for_message is provided by the Mcp Switchboard MCP server (jemplayer82/mcp-switchboard). 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 →