image_compare_compare · Leaper Vision Toolkit MCP

WHAT IT DOES

What image_compare_compare does on Leaper Vision Toolkit

AI agents call image_compare_compare to retrieve information from Leaper Vision Toolkit without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

PARAMETERS

Parameter	Type	Required	Description
`imageUrl`	string	—	图片的url地址，通过这个地址获取到输入图像
`regionClassObjDescriptionJson`	string	—	原始的Roi区域。在调用此工具前要保证前处理中一定要有且仅有一次的 Region 初始化工具调用，可用的初始化工具有：ILAnnulusRegion (圆环区域对象)、ILAnnulusSectorRegion(环状扇形区域对象)、ILCircleRegion (圆形区域对象)、ILCompoundRegion (复合
`imageCompareClassObjDescriptionJson`	string	—	用于图像比较功能调用的对象。在调用此工具前要保证前处理中一定要有且仅有一次的 ImageCompareNew 初始化工具调用。如果用户指定图像比较功能相关对象，传入一个序列化的JSON字符串，属性包括：LpvClassName(调用工具函数的名称) 和 InputParameterFile(调用工具函数需要传入的参数文

Parameters from the server's own tool schema.

WHY IT NEEDS A POLICY

Why image_compare_compare needs a policy

This tool performs a read/analysis operation: it compares an input image against a baseline and returns difference results. It does not create, modify, delete, or execute anything. The outputs (diffResult, maskResult) are computed results returned to the caller, not stored changes.

From the tool's definition Compare with the baseline image (and mask). The tool reads an input image and region, then outputs a difference image and mask — no data is modified or deleted.

Attacks that exploit this kind of access

Questions about image_compare_compare

What does the image_compare_compare tool do? +

Compare with the baseline image (and mask). * @param[in] img The input image * @param[in] region The input region, to limit the region of the comparison on the input image * and also exclude some not-care pixels. * @param[out] diffResult The pixel-level difference image in the same format as the input image * and same size as the input region (or the input image if the region is empty) * @param[out] maskResult @bin_image @bit8 The pixel-level difference mask, white pixels represent those differences exceed #PixelTolerance 与基准图像（和蒙版）进行比较。 * @param[in] img 输入的图像 * @param[in] region 输入的区域，用于限定比较区域在输入图像中的位置，和剔除无关像素 * @param[out] diffResult 像素差异图，图像类型与输入图像相同，尺寸与输入区域相同（若输入区域为空，则与输入图像相同） * @param[out] maskResult @bin_image @bit8 像素差异蒙版图，白色像素表示差异大于 #PixelTolerance 与基准图像（和蒙版）进行比较。没有设置基准图像时，需要让用户先通过 ImageCompareSetBaseline 设置基准图像。返回结构是一个序列化的json，其中： DiffResultImg属性是差异图像URL地址， MaskResultImg属性是差异蒙版图像URL地址. It is categorised as a Read tool in the Leaper Vision Toolkit MCP Server, which means it retrieves data without modifying state.

What parameters does image_compare_compare accept? +

image_compare_compare accepts 3 parameters: imageUrl, regionClassObjDescriptionJson, imageCompareClassObjDescriptionJson. The full parameter table on this page comes from the server's own tool schema.

How do I enforce a policy on image_compare_compare? +

Register the Leaper Vision Toolkit MCP server in PolicyLayer and add a rule for image_compare_compare: 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 Leaper Vision Toolkit. Nothing to install.

What risk level is image_compare_compare? +

image_compare_compare is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit image_compare_compare? +

Yes. Add a rate_limit block to the image_compare_compare 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.

How do I block image_compare_compare completely? +

Set action: deny in the PolicyLayer policy for image_compare_compare. 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.

What MCP server provides image_compare_compare? +

image_compare_compare is provided by the Leaper Vision Toolkit MCP server (leaper-mcp/leaper-mcp-proxy). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.