AI agents call vision_assistant to retrieve information from xigua-MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The description translates to 'Visual Perception System', suggesting this tool reads/captures visual data from the environment (e.g., camera, screen). It likely retrieves visual information rather than modifying anything.
From the tool's definition Tool name 'vision_assistant' and description '视觉感知系统' (Visual Perception System)
Attacks that exploit this kind of access
视觉感知系统. It is categorised as a Read tool in the xigua-MCP MCP Server, which means it retrieves data without modifying state.
Register the xigua- MCP server in PolicyLayer and add a rule for vision_assistant: 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 xigua-MCP. Nothing to install.
vision_assistant 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 vision_assistant 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 vision_assistant. 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.
vision_assistant is provided by the xigua- MCP server (xiguaxiaome/xigua-mcp). 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 →