Leaper Vision Toolkit MCP Policy · 169 Tools

Q: How do I prevent bulk modifications through Leaper Vision Toolkit?

The Leaper Vision Toolkit server has 113 write tools including edge_filter_gradient_full, edge_filter_gradient_h, edge_filter_set_edge_kernel. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Leaper Vision Toolkit.

Q: How many tools does the Leaper Vision Toolkit MCP server expose?

169 tools across 4 categories: Destructive, Execute, Read, Write. 31 are read-only. 138 can modify, create, or delete data.

Q: How do I enforce a policy on Leaper Vision Toolkit?

Register the Leaper Vision Toolkit MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

POLICY

How to control Leaper Vision Toolkit

PolicyLayer is an MCP gateway — it sits between your AI agents and Leaper Vision Toolkit, and nothing reaches the server without passing your rules. These are the rules we recommend:

Rate limit write operations

{
  "edge_filter_gradient_full": {
    "limits": [
      {
        "counter": "edge_filter_gradient_full_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "display_blob_analysis": {
    "limits": [
      {
        "counter": "display_blob_analysis_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Leaper Vision Toolkit — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON LEAPER VISION TOOLKIT →

Instant setup, no code required.

FULL CATALOGUE

All 169 Leaper Vision Toolkit tools

EXECUTE 25 tools

Execute display_match Display the pattern matching results. 这个是执行模板匹配的图像的交互工具，并返回标记斑点后的图片。 This interface provide functionalities o Execute edge_filter_gradient_v Vertically oriented edges are represented by bright pixels. 垂直梯度，垂直边缘在输出图像中表现为较亮的像素。 Provides image edge enh Execute edge_filter_new This interface provide functionalities of edge extraction. 该接口类提供了图像边缘增强的功能。 To use this interface, you shou Execute il_blob_analysis This interface provide functionalities of blob analysis. 该接口提供斑点分析的功能。 Blob analysis is a useful approach to Execute il_blob_analysis_color This interface provide functionalities of blob analysis. 该接口提供斑点分析的功能。 Blob analysis is a useful approach to Execute il_image_filter_high_pass 高通滤波工具，相对于梯度更一致的区域，增强图像中的边。 Apply a high-pass filter to the input image to enhance edges. High-pass filter e Execute il_image_filter_linear_filter_abs 基于自定义核，对输入图像进行线性滤波，输出滤波结果的绝对值。 Apply linear filtering on the given image using the custom kernel. kMat then Execute il_image_filter_local_median_norm 归一化中值滤波，用于从图像中去除背景像素。该滤波器相当于输入图像与中值滤波的结果相减，然后将每个像素 << 1 + 128。 Used to subtract the background from an image Execute il_image_filter_optical_density 光密度，通过度量其通光量显露在输入图像中物体或特征的密度。密度较高的物体或特征在输出图像中以较亮的像素表示。 Reveals the density of objects and features in the i Execute il_image_morph_gradient 这个是梯度工具。输出形态学意义上的梯度图像。该工具等价于膨胀减去腐蚀的结果。用户没有指定形态学相关参数时，不要调用设置参数的插件，传入的 ILImageMorph 类的文件中已经设置了形态学相关的默认参数。返回结 Execute il_image_morph_top_bottom_hat 这个是高低帽工具。每个像素输出高帽结果或低帽结果中的最大值。用户没有指定形态学相关参数时，不要调用设置参数的插件，传入的 ILImageMorph 类的文件中已经设置了形态学相关的默认参数。返回结构是一个序列化的 Execute il_image_op_diff_scalar 这个是计算输入图像和一个标量逐像素差值的工具。输出一个新图像。工具不需要通过其它工具设置参数，不需要传入ILImageOp的对象。返回结构是一个序列化的json，其中： ResultImg属性是输出结果图像U Execute il_image_op_normalize_mean_std_dev 这个是归一化输入图像，以当前图像的均值和标准差计算感兴趣的数值范围，拉伸到目标数值范围的工具。不改变图像位深，归一化后的结果图像与输入位深相同、通道数相同。对于彩色的多通道图像，输出也是彩色多通道图像，每个通道分别独立进 Execute il_image_op_scale_rotate 这个是缩放和旋转输入图像的工具。使用用户指定的锚点。若指定的结果图像对象为空图像，则生成和输入图像尺寸相同的图像。否则，将保持结果图像的尺寸，超出图像范围的像素设为背景像素（默认为黑色）。工具不需要通过其它工具设置参 Execute il_image_op_split 这个是按网格拆分一个图像的工具。为若干个小图。工具不需要通过其它工具设置参数，不需要传入ILImageOp的对象。返回结构是一个序列化的json，其中： ResultImgList属性是输出结果图像URL地 Execute il_image_op_split_x 这个是水平拆分一个图像的工具。为若干个小图。工具不需要通过其它工具设置参数，不需要传入ILImageOp的对象。返回结构是一个序列化的json，其中： ResultImgList属性是输出结果图像URL地址 Execute il_image_threshold LPV Image Threshold initialization tool. This is the initialization tool for image threshold operations. It is Execute il_image_threshold_binarize Binarize the input image and return the result new image. If the lower bound of threshold value is smaller tha Execute il_image_threshold_stretch Stretch the input image and output the result image. Each output pixel represents the difference between the g Execute il_region_rotate_and_translate Rotate and translate a region based on anchor point transformation. 基于锚点的位置和角度变化，生成一个旋转和平移后的ROI区域对象。 The rela Execute il_region_scale Scale a region while keeping the center point unchanged. 缩放当前ROI区域，生成一个新的ROI区域对象，保持区域中心点不变。 The region is sca Execute il_region_translate Translate a region by the specified offset. 平移当前ROI区域，生成一个新的ROI区域对象。 The region is moved by the specified X a Execute image_convert_bgr_to_gray_color 将 BGR 颜色值转换为灰度值返回结构是一个序列化的JSON字符串，包含以下字段： GrayValue: 转换后的灰度值 Execute image_convert_convert Convert the color image from one color space to another, for example, from BGR to HSV 对彩色图像进行颜色空间转换，如，从 BGR 转 Execute pat_match_new Create a new pattern matching tool instance. 创建一个新的模板匹配工具实例。 This interface provides functionalities of patt

WRITE 113 tools

READ 31 tools

Read display_blob_analysis Display the blob analysis results. 这是针对斑点的图像的交互工具，并返回标记斑点后的图片。 This interface provide functionalities of the Read display_circle_detector Display the circle detection results. 这个是针对圆定位的图像的交互工具，并返回检测和定位圆后的图片。 This interface provide functionalities Read display_line_detector Display the line detection results. 这个是针对直线定位的图像的交互工具，并返回检测和定位直线后的图片。 This interface provide functionalities Read get_tool_tree 获取工具节点和下一层子节点的工具名称的方法，每次调用仅获取一层子节点，主要用于列出某个模块的方法和获取所有方法列表。参数说明: - name: 要查询的节点名称，并根据name返回该节点及其下一层节点，不传则返回根节点 Read il_histogram_build_draw Build histogram from input image and region, and draw as image. Draw the histogram as a data chart onto provi Read il_histogram_build_items Build histogram from input image and region, output the pixel count for each bin. Collects count of data in p Read il_histogram_build_mean_std_dev Build histogram from input image and region, calculate mean and standard deviation. Calculate the mean and st Read il_histogram_build_min_max Build histogram from input image and region, find minimum and maximum values. Find the bin of minimum or maxi Read il_image_op_project_x 这个是水平投影图像的工具。聚合每一行上的像素。工具不需要通过其它工具设置参数，不需要传入ILImageOp的对象。返回结构是一个序列化的json，其中： ResultArray属性是输出的结果向量列表。 Read il_image_op_project_y 这个是垂直投影图像的工具。聚合每一列上的像素。工具不需要通过其它工具设置参数，不需要传入ILImageOp的对象。返回结构是一个序列化的json，其中： ResultArray属性是输出的结果向量列表。 Read ilocr_ib_detect 这个是分割并识别行和字符工具。基于输入的图像和区域,分割并识别行和字符,返回字符分割和识别的结果和分数。Results 是字符分割和识别的信息，其中 ocrRes 表示分割和识别的所有结果，每个字符用空格隔开。ocrSc Read image_compare_compare Compare with the baseline image (and mask). @param[in] img The input image @param[in] region The input reg Read image_compare_compare_symmetric 比较输入图像中的两个对称区域，生成差异图和蒙版图。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 Read image_compare_new This interface provide functionalities of image comparison on pixel level. To use this interface, you should Read image_convert_gray_to_bgr_color 将灰度值转换为 BGR 颜色值返回结构是一个序列化的JSON字符串，包含以下字段： Blue: 蓝色通道值 Green: 绿色通道值 Red: 红色通道值 Read image_get_channel BGR8 从彩色图像中提取某一个给定索引的通道图像。仅用于彩色图像。返回结构是一个序列化的JSON字符串，包含以下字段： ResultImg: 给定索引的通道的图像结果 Extract the channe Read image_info 获取输入图像的图片信息。返回结构是一个序列化的JSON字符串，包含以下字段： BytesPerLine : 每行像素数据的字节数，包含行末用于对齐的字节。 Height: 图像的高度，行的总数。 W Read image_stats_count_pixel Count the pixels in the input image which is below, between or above the given threshold bounds. 计算输入图像中灰度值在 Read image_stats_count_pixel_color Count the pixels in the input image which is between or outside the given color bounds. For all of the channe Read image_stats_max Find the pixel of maximum value in the input image, report the value and pixel's position. If there's more th Read image_stats_mean_std_dev Calculate the mean and standard deviation of the input image. 计算输入图像的灰度均值和标准差。 Returns a JSON string with th Read image_stats_min Find the pixel of minimum value in the input image, report the value and pixel's position. If there's more th Read image_stats_sharpness Compute the relative sharpness of the input image, based on the edge gradients. It's usually used to automate Read pat_match_get_pat_center Get the center of the pattern template. 获取当前配置的模板中心。 The template center indicates the report position of the Read pat_match_get_pat_center_mode Get the center mode of the pattern template. 获取当前配置的模板中心设置方式。 This enumeration represents the mode of the pat Read pat_match_get_pat_feature Get the feature points of trained pattern. 获取模板的特征点。 The feature points are extracted from the trained templa Read pat_match_get_pat_image Get the pattern template image. 获取模板图像。 The template image is the original image used for training the patter Read pat_match_get_pat_mask Get the pattern template mask. 获取模板蒙版。 The mask is used to exclude some not-care pixels during training. 蒙版用于 Read pat_match_get_pat_prune_mask Get the prune mask. 获取特征点剔除蒙版。 The prune mask shows which features have been excluded from the pattern. 特征点剔除 Read pat_match_get_pat_shape Get the shape of trained pattern as polylines. 获取模板的形状线。 The shape is represented as polylines extracted from Read pat_match_is_learnt Check whether the LMatch object is well-trained. 检查当前模板匹配的功能对象是否已训练。 Return True if it's trained, otherwise,

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Leaper Vision Toolkit

How do I prevent bulk modifications through Leaper Vision Toolkit? +

The Leaper Vision Toolkit server has 113 write tools including edge_filter_gradient_full, edge_filter_gradient_h, edge_filter_set_edge_kernel. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Leaper Vision Toolkit.

How many tools does the Leaper Vision Toolkit MCP server expose? +

169 tools across 4 categories: Destructive, Execute, Read, Write. 31 are read-only. 138 can modify, create, or delete data.

How do I enforce a policy on Leaper Vision Toolkit? +

Register the Leaper Vision Toolkit MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Leaper Vision Toolkit

What Leaper Vision Toolkit exposes to your agents

What Leaper Vision Toolkit costs in tokens

The most dangerous Leaper Vision Toolkit tools

How to control Leaper Vision Toolkit

All 169 Leaper Vision Toolkit tools

Related servers

Questions about Leaper Vision Toolkit

Enforce policy on every Leaper Vision Toolkit tool call.