← Back to Skills Marketplace
chengjunai

通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等

by chengjunai · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
103
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install chengjun-content-security
Description
调用成均平台HTTP API,检测文本中的敏感词、领导官员称谓、禁用词、语法和标点错误,保障内容合规安全。
README (SKILL.md)

成均内容安全 Skill

name: 成均内容安全 description: 通过调用成均平台的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等,保障内容合规。

简介

这是一个用于检测文本敏感信息的 Skill,通过调用成均智能平台(博达智能体开放平台)的内容安全接口,快速识别文本内容中的敏感数据、语法错误等问题,适用于内容审核、数据合规等场景。

本技能支持检测以下内容:

  • 敏感词汇
  • 领导人称谓
  • 官员称谓
  • 禁用词汇
  • 语法错误
  • 标点符号问题

⚠️ 重要配置说明

API 认证配置

方式:直接使用Token

export CHENGJUN_API_KEY="your_http_api_token"

服务信息

功能特性

  • 支持多种内容安全检查(敏感词、领导人称谓、官员称谓、禁用词、语法词、标点符号)
  • 支持直接Token和自动获取Token两种认证方式
  • Token自动缓存管理,避免频繁请求
  • 标准化的 JSON 格式输入输出
  • 完善的错误处理和超时机制
  • 兼容 OpenClaw 对话式调用
  • 文本长度限制:5000 字符
  • 使用 Python 标准库,无需额外依赖

安全说明

API 认证

  • 支持直接传入 Token
  • Token 通过 Authorization 请求头传递

数据安全

  • 预检结果不会被服务端保存
  • 单次请求文本长度不超过 5000 字符
  • 使用 HTTPS 协议传输数据

技术实现

使用的技术栈

技术

用途

Python 3.8+

主要开发语言

urllib (标准库)

HTTP 请求库

依赖安装

本 skill 使用 Python 标准库,无需安装额外依赖。

使用方法

对话式调用

请帮我检测这段文本是否包含敏感信息:{待检测文本内容}

API 认证配置

方式:直接使用Token

export CHENGJUN_API_KEY="your_http_api_token"

参数说明

参数名

类型

必传

说明

text

string

待检测的文本内容(不超过5000字符)

api_key

string

API密钥(可选,默认从环境变量读取)

返回格式

{
  "code": 200,
  "msg": "检测成功",
  "precheck": {
    "leader": [],
    "officer": [],
    "forbidden": [],
    "grammar_word": [],
    "sensitive": [
      {
        "word": "党的二十",
        "correct": ["党的二十大"],
        "misdescription": null,
        "category": "中国共产党相关表述",
        "sentence": "为深入学习贯彻党的二十精神",
        "explain": null,
        "startInSentence": 9,
        "riskLevel": "middle",
        "level": "疑似错误",
        "start": 9,
        "end": 13,
        "priority": 2,
        "typeNum": 1
      }
    ],
    "punctuation": [],
    "urlBody": null,
    "wordNum": 481
  }
}

检查项说明

检查项

说明

leader

领导人称谓检查结果

officer

官员称谓检查结果

forbidden

禁用词检查结果

grammar_word

语法词检查结果

sensitive

敏感词检查结果

punctuation

标点符号检查结果

错误码说明

通用错误码

错误码

说明

200

检测成功

-1

业务逻辑错误(文本为空或超长)

API 调用错误

错误码

说明

401

认证失败(Token无效或凭证错误)

403

权限不足

429

请求频率超限

500

服务器内部错误

-2

API 调用失败(网络或服务问题)

示例代码

from main import check_content, run

# 方式1:使用环境变量中的Token
# export CHENGJUN_API_KEY="your_http_api_token"
result = check_content(
    text="创建特色社会主义道路,为深入学习贯彻党的二十精神"
)

# 方式2:直接传入Token
result = check_content(
    text="创建特色社会主义道路,为深入学习贯彻党的二十精神",
    api_key="your_http_api_token"
)

# 使用run函数(兼容OpenClaw)
result = run({
    "text": "创建特色社会主义道路,为深入学习贯彻党的二十精神"
})

# 打印结果
print(json.dumps(result, ensure_ascii=False, indent=2))

注意事项

  1. 使用前需要配置API认证(Token)
  2. 单次请求文本长度不超过 5000 字符
  3. 预检结果不会被服务端保存
  4. 接口调用超时时间为 30 秒
  5. Token自动缓存,有效期内的请求不会重复获取

API 接口文档

详细文档请参考:https://platform.vsbclub.com/docs/zh/webber/content_check.html

文本预检接口

故障排查

如果调用失败,检查:

  1. 网络连接curl -I https://api.vsbclub.com/
  2. Token配置:确认 CHENGJUN_API_KEY 设置正确
  3. 文本长度:确保不超过5000字符
  4. 服务状态:服务可能暂时不可用,稍后重试

获取 API Token

直接获取可用的 HTTP API Token https://platform.vsbclub.com/

Usage Guidance
Review carefully before installing. The integration itself is purpose-aligned, but the current code should not be used for sensitive or production content until HTTPS certificate verification is re-enabled and the handling of submitted text is clarified.
Capability Analysis
Type: OpenClaw Skill Name: chengjun-content-security Version: 1.0.1 The skill contains a significant security vulnerability in main.py where it globally disables SSL certificate verification (ssl._create_unverified_context), which exposes the agent to Man-in-the-Middle (MITM) attacks. Additionally, main.py contains redundant and poorly implemented code, including duplicate function definitions for check_content() that reference undefined variables (e.g., TOKEN_URL, get_api_credentials). While the behavior aligns with the stated purpose of calling the vsbclub.com API for content moderation, the intentional weakening of transport security and poor code quality warrant a suspicious classification.
Capability Tags
requires-oauth-tokenrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The skill’s stated purpose, API endpoint, text-length limit, and output format are coherent with a content-safety checking integration.
Instruction Scope
The skill sends user-provided text to an external provider for checking; this is disclosed and purpose-aligned, but users should avoid submitting highly sensitive content unless they trust the provider.
Install Mechanism
There is no install script and no third-party dependency, but the registry metadata says no required environment variables while the skill.json and documentation require CHENGJUN_API_KEY.
Credentials
The code globally disables SSL certificate verification before sending the API key and text to the provider, weakening the claimed HTTPS protection.
Persistence & Privilege
No filesystem persistence, background process, or long-running autonomous behavior is shown; the API key cache is in-memory only.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install chengjun-content-security
  3. After installation, invoke the skill by name or use /chengjun-content-security
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- 简化和精简了 API 认证说明,仅保留“直接使用Token”方式 - 移除了通过凭证自动获取Token相关内容及说明 - 更新 API 文档与服务说明,统一指向平台入口 https://platform.vsbclub.com/ - 精简了故障排查和注意事项,删除与凭证获取相关的指引 - 文档内容整体更为清晰聚焦,只保留和 Token 配置相关的内容
v1.0.0
Initial release of the 成均内容安全 Skill: - Supports content security checks for sensitive words, leader and officer titles, forbidden words, grammar, and punctuation. - Offers two API authentication methods: direct token or credential-based token retrieval. - Provides standardized JSON input/output, robust error handling, and token auto-caching. - Uses Python standard library only; no extra dependencies required. - Compatible with OpenClaw dialog-style invocation. - Enforces a 5000-character text length limit per request. - Comprehensive usage documentation and sample code included.
Metadata
Slug chengjun-content-security
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is 通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等?

调用成均平台HTTP API,检测文本中的敏感词、领导官员称谓、禁用词、语法和标点错误,保障内容合规安全。 It is an AI Agent Skill for Claude Code / OpenClaw, with 103 downloads so far.

How do I install 通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等?

Run "/install chengjun-content-security" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等 free?

Yes, 通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等 support?

通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 通过调用成均智能的HTTP API接口,检测文本中的敏感信息、语法错误、标点符号问题等?

It is built and maintained by chengjunai (@chengjunai); the current version is v1.0.1.

💬 Comments