Check for new messages from the AI Companion panel in the Chrome extension. Returns pending messages and clears the queue. Each message includes the user text, page context, and optional screenshot/element attachments.
AI agents call get_companion_messages to retrieve information from Yocoolab without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves messages from an AI Companion panel queue. While it clears the queue after reading, this is a transient state management operation consistent with message polling—the messages themselves are not modified, deleted irreversibly, or used to trigger external actions. The blast radius of misuse is confined to reading companion messages already in the queue, posing minimal risk.
From the tool's definition Tool description states it 'Check[s] for new messages' and 'Returns pending messages'. The operation retrieves data (messages from a queue) without modifying, deleting, or executing external operations.
Attacks that exploit this kind of access
Check for new messages from the AI Companion panel in the Chrome extension. Returns pending messages and clears the queue. Each message includes the user text, page context, and optional screenshot/element attachments. It is categorised as a Read tool in the Yocoolab MCP Server, which means it retrieves data without modifying state.
Register the Yocoolab MCP server in PolicyLayer and add a rule for get_companion_messages: 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 Yocoolab. Nothing to install.
get_companion_messages 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 get_companion_messages 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 get_companion_messages. 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.
get_companion_messages is provided by the Yocoolab MCP server (yocoolab/mcp-server). 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 →