AI agents call channel_read to retrieve information from Wmux without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves message data from accessible channels without modifying, deleting, or executing any operations. It is a passive read operation analogous to fetching or querying stored messages. The severity is low because reading messages from channels the agent is authorized to access presents minimal risk; the blast radius is limited to information disclosure of already-accessible data.
From the tool's definition Tool name 'channel_read' and description explicitly states 'Read messages from a channel' — a retrieval operation with no side effects.
Attacks that exploit this kind of access
Read messages from a channel you can see (public, or private if you are a member). It is categorised as a Read tool in the Wmux MCP Server, which means it retrieves data without modifying state.
Register the Wmux MCP server in PolicyLayer and add a rule for channel_read: 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 Wmux. Nothing to install.
channel_read 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 channel_read 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 channel_read. 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.
channel_read is provided by the Wmux MCP server (openwong2kim/wmux). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
channel_read is one line of Wmux's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →