Answer a question about an image's contents using state-of-the-art generative AI models.
AI agents call query_image to retrieve information from Perception-MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs analysis and retrieval of information from image files without creating, modifying, deleting, or executing code. It reads image data and returns analytical results. No side effects or state changes occur. This is consistent with the Read category for tools that retrieve or query data without side effects.
From the tool's definition Tool description states it will 'Answer a question about an image's contents' - a retrieval/analysis operation with no modification of data or execution of external actions. The verb 'Answer' and 'query' indicate information retrieval only.
Attacks that exploit this kind of access
Answer a question about an image's contents using state-of-the-art generative AI models. It is categorised as a Read tool in the Perception-MCP MCP Server, which means it retrieves data without modifying state.
Register the Perception- MCP server in PolicyLayer and add a rule for query_image: 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 Perception-MCP. Nothing to install.
query_image 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 query_image 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 query_image. 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.
query_image is provided by the Perception- MCP server (lintyourcode/perception-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 →