Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

TencentCloud RecognizeTable OCR

v1.0.2

腾讯云表格识别v3(RecognizeTableAccurateOCR)接口调用技能。当用户需要从表格图片或PDF中识别常规表格、无线表格、多表格的内容,提取每个单元格的文字信息,或将表格图片识别结果导出为Excel文件时,应使用此技能。支持中英文表格图片、旋转表格图片、嵌套表格图片等复杂场景,识别效果优于表格识...

0· 516·1 current·1 all-time
bytencent-ocr@zt1314p-design

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for zt1314p-design/tencentcloud-ocr-recognizetableaccurate.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "TencentCloud RecognizeTable OCR" (zt1314p-design/tencentcloud-ocr-recognizetableaccurate) from ClawHub.
Skill page: https://clawhub.ai/zt1314p-design/tencentcloud-ocr-recognizetableaccurate
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Canonical install target

openclaw skills install zt1314p-design/tencentcloud-ocr-recognizetableaccurate

ClawHub CLI

Package manager switcher

npx clawhub@latest install tencentcloud-ocr-recognizetableaccurate
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill's code and SKILL.md implement Tencent Cloud RecognizeTableAccurateOCR and legitimately require Tencent Cloud API credentials and the tencentcloud-sdk-python dependency. However, the registry metadata lists no required environment variables or primary credential, which is inconsistent with the declared purpose.
Instruction Scope
Runtime instructions are scoped to calling the Tencent Cloud OCR API. The script only reads the image/PDF provided via URL or a user-supplied file path (or Base64 string) and transmits that content to Tencent Cloud. It does not attempt to read other system files or additional environment variables.
Install Mechanism
There is no install spec in the registry; the SKILL.md tells users to pip install tencentcloud-sdk-python. Leaving dependency installation to the user is reasonable but means the skill will fail unless the SDK is installed; no additional installers or remote downloads are present in the bundle.
!
Credentials
The script requires TENCENTCLOUD_SECRET_ID and TENCENTCLOUD_SECRET_KEY (which is proportional to calling Tencent's API). The concern is the metadata (registry) omits those required environment variables and primary credential — this metadata mismatch can hide the need for secrets and mislead users. No unrelated secrets are requested by the code.
Persistence & Privilege
The skill does not request persistent/always-on privileges, does not modify other skills or system configs, and only runs when invoked. It does read and may write a user-specified file path for saving Excel output (expected behavior).
What to consider before installing
This skill's code matches its description — it sends images/PDFs to Tencent Cloud OCR and requires your Tencent Cloud API key and secret. Before installing or running it: 1) Be aware that image/PDF contents will be transmitted to Tencent Cloud (do not send sensitive documents unless you accept that). 2) The registry metadata incorrectly omits the required env vars; ensure you or the deployment environment provide TENCENTCLOUD_SECRET_ID and TENCENTCLOUD_SECRET_KEY. 3) Use a least-privileged Tencent API key (create a key scoped only to the OCR service if possible) and run the script in an isolated environment. 4) Install the tencentcloud-sdk-python package from PyPI yourself (pip install tencentcloud-sdk-python) rather than trusting any automated installer. 5) If you need higher assurance, ask the publisher to fix the metadata to list required credentials or request a signed/official release location.

Like a lobster shell, security has layers — review code before you run it.

latestvk9762w3wvnymb4h8nxg0msek6582n2m8
516downloads
0stars
3versions
Updated 19h ago
v1.0.2
MIT-0

腾讯云表格识别v3 (RecognizeTableAccurateOCR)

用途

调用腾讯云OCR表格识别v3接口,支持中英文图片/PDF内常规表格、无线表格、多表格的检测和识别,返回每个单元格的文字内容,支持旋转的表格图片识别,且支持将识别结果保存为Excel格式。

核心能力:

  • 常规表格识别:支持有线表格的精准识别
  • 无线表格识别:支持无边框表格的检测与识别
  • 多表格识别:单张图片/PDF中包含多个表格时均可检测
  • 嵌套表格识别:有线表格中包含无线表格的复杂场景
  • 旋转表格识别:支持旋转角度的表格图片
  • Excel导出:识别结果可直接导出为Excel文件(Base64编码)
  • PDF支持:支持PDF文件的单页表格识别

官方文档:https://cloud.tencent.com/document/api/866/86721

使用时机

当用户提出以下需求时触发此技能:

  • 需要从图片中提取表格内容
  • 需要从PDF中识别表格数据
  • 需要将图片/PDF中的表格转换为Excel文件
  • 涉及表格OCR识别的任何场景
  • 需要识别无线表格、嵌套表格等复杂表格

环境要求

  • Python 3.6+
  • 依赖:tencentcloud-sdk-python(通过 pip install tencentcloud-sdk-python 安装)
  • 环境变量:
    • TENCENTCLOUD_SECRET_ID:腾讯云API密钥ID
    • TENCENTCLOUD_SECRET_KEY:腾讯云API密钥Key

使用方式

运行 scripts/main.py 脚本完成表格识别。

请求参数

参数类型必填说明
ImageBase64str否(二选一)图片/PDF的Base64值,编码后不超过10M,分辨率建议600*800以上且长宽比小于3,支持PNG/JPG/JPEG/BMP/PDF
ImageUrlstr否(二选一)图片/PDF的URL地址,都提供时只使用ImageUrl
PdfPageNumberintPDF页码,仅支持单页识别,默认1,必须>=1
UserAgentstr请求来源标识(可选),用于追踪调用来源,统一固定为Skills

输出格式

识别成功后返回 JSON 格式结果:

格式化输出模式(默认)

{
  "表格数量": 1,
  "表格详情": [
    {
      "表格序号": 1,
      "表格类型": "有线表格",
      "单元格数量": 12,
      "单元格详情": [
        {
          "行范围": "0 - 1",
          "列范围": "0 - 1",
          "文本": "本报告期末",
          "置信度": 99.0,
          "单元格类型": "body"
        }
      ],
      "表格坐标": [
        {"X": 50, "Y": 100},
        {"X": 800, "Y": 100},
        {"X": 800, "Y": 600},
        {"X": 50, "Y": 600}
      ]
    }
  ],
  "Excel数据": "已返回(Base64编码)",
  "旋转角度": 0.0,
  "RequestId": "xxx"
}

原始输出模式(--raw)

{
  "TableDetections": [
    {
      "Cells": [
        {
          "ColTl": 0,
          "RowTl": 0,
          "ColBr": 1,
          "RowBr": 1,
          "Text": "本报告期末",
          "Type": "body",
          "Confidence": 99.0,
          "Polygon": [{"X": 50, "Y": 100}, {"X": 200, "Y": 100}, {"X": 200, "Y": 150}, {"X": 50, "Y": 150}]
        }
      ],
      "Type": 1,
      "TableCoordPoint": [{"X": 50, "Y": 100}, {"X": 800, "Y": 100}, {"X": 800, "Y": 600}, {"X": 50, "Y": 600}]
    }
  ],
  "Data": "UEsDBBQACAgIAIFzWFY...",
  "PdfPageSize": 0,
  "Angle": 0.0,
  "RequestId": "xxx"
}

响应数据结构说明

TableInfo 结构:

字段类型说明
Cellslist of TableCellInfo单元格内容(可能返回null)
Typeint文本块类型:0非表格文本、1有线表格、2无线表格(可能返回null)
TableCoordPointlist of Coord表格主体四个顶点坐标,左上→右上→右下→左下(可能返回null)

TableCellInfo 结构:

字段类型说明
ColTlint单元格左上角的列索引
RowTlint单元格左上角的行索引
ColBrint单元格右下角的列索引
RowBrint单元格右下角的行索引
Textstr单元格内识别出的字符串文本,多行以换行符\n隔开
Typestr单元格类型
Confidencefloat单元格置信度
Polygonlist of Coord单元格在图像中的四点坐标

错误码说明

错误码含义
FailedOperation.DownLoadError文件下载失败
FailedOperation.EmptyImageError图片内容为空
FailedOperation.ImageDecodeFailed图片解码失败
FailedOperation.ImageSizeTooLarge图片尺寸过大
FailedOperation.OcrFailedOCR识别失败
FailedOperation.PDFParseFailedPDF解析失败
FailedOperation.UnKnowError未知错误
FailedOperation.UnKnowFileTypeError未知的文件类型
FailedOperation.UnOpenError服务未开通
InvalidParameterValue.InvalidParameterValueLimit参数值错误
LimitExceeded.TooLargeFileError文件内容太大
ResourceUnavailable.InArrears账号已欠费
ResourceUnavailable.ResourcePackageRunOut账号资源包耗尽
ResourcesSoldOut.ChargeStatusException计费状态异常

重要业务逻辑

  1. ImageBase64和ImageUrl必须提供其一,都提供时只使用ImageUrl
  2. 图片/PDF编码后不超过10M,分辨率建议600800以上且长宽比小于3(短边分辨率>600,长边<=短边3)
  3. 支持格式:PNG、JPG、JPEG、BMP、PDF
  4. PdfPageNumber必须>=1,仅支持PDF单页识别,默认值为1
  5. PDF转图有白名单机制,需白名单用户才支持
  6. 若引擎返回ImageSizeTooLargeError,会映射为ImageSizeInvalidError
  7. 支持将识别结果导出为Excel(Data字段为Base64编码的Excel数据)
  8. 默认接口请求频率限制:2次/秒
  9. 建议图片存储于腾讯云COS以获得更高的下载速度和稳定性

调用示例

# 通过URL识别表格
python scripts/main.py --image-url "https://example.com/table.jpg"

# 通过文件路径(自动Base64编码)识别
python scripts/main.py --image-base64 ./table.png

# 识别PDF中的表格(指定页码)
python scripts/main.py --image-base64 ./document.pdf --pdf-page-number 2

# 识别并保存Excel文件
python scripts/main.py --image-url "https://example.com/table.jpg" --save-excel ./result.xlsx

# 输出原始JSON响应
python scripts/main.py --image-url "https://example.com/table.jpg" --raw

# 指定地域
python scripts/main.py --image-url "https://example.com/table.jpg" --region ap-beijing```

Comments

Loading comments...