AI agents call lens_watch to retrieve information from Lens without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries/retrieves annotation data with no side effects on the system state. It blocks waiting for new data but does not create, modify, delete, or execute external operations. The context of a visual feedback and QA assertions system suggests this is a monitoring/observation function.
From the tool's definition Tool description states it 'Long-poll for new pending annotations' — a blocking read operation that retrieves incoming annotations without modifying them. The verb 'poll' and 'for' indicate data retrieval only.
Attacks that exploit this kind of access
Long-poll for new pending annotations (blocks until new annotations arrive). It is categorised as a Read tool in the Lens MCP Server, which means it retrieves data without modifying state.
Register the Lens MCP server in PolicyLayer and add a rule for lens_watch: 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 Lens. Nothing to install.
lens_watch 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 lens_watch 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 lens_watch. 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.
lens_watch is provided by the Lens MCP server (renderdraw/lens). 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 →