AI agents call human_eye_tool to retrieve information from Human without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries human sensory perception to obtain information about the physical world. It is analogous to a search or fetch operation, returning observational data without side effects. No data is modified, deleted, or financial transactions occur. The tool cannot execute code or trigger external operations beyond requesting human observation.
From the tool's definition Tool description (Japanese): '人間が目で見て状況を説明したり、特定のものを探したりします' translates to 'A human looks with their eyes to describe situations or search for specific things.' The tool retrieves visual information and descriptions from a human observer—a pure…
Documented attack patterns abuse exactly the kind of access human_eye_tool gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Human, and nothing reaches the server without passing your rules. This is the rule we recommend for human_eye_tool:
{
"version": "1",
"default": "deny",
"tools": {
"human_eye_tool": {}
}
}human_eye_tool is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
人間が目で見て状況を説明したり、特定のものを探したりします。. It is categorised as a Read tool in the Human MCP Server, which means it retrieves data without modifying state.
Register the Human MCP server in PolicyLayer and add a rule for human_eye_tool: 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 Human. Nothing to install.
human_eye_tool 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 human_eye_tool 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 human_eye_tool. 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.
human_eye_tool is provided by the Human MCP server (upamune/human-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 7 Human tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
7 Human tools catalogued and risk-classified — across an index of 42,500+ MCP servers.