Skill flagged — suspicious patterns detected

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

AI Customer Service Automation

24/7智能客服系统,自动回复常见问题,情感分析升级工单,多渠道支持,数据看板和知识库同步。

MIT-0 · Free to use, modify, and redistribute. No attribution required.
0 · 91 · 0 current installs · 0 all-time installs
MIT-0
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The SKILL.md describes multi-channel support (Feishu/WeChat/DingTalk/Telegram/email), Jira ticketing, sentiment analysis, KB sync and gives a config example — that matches the declared purpose. However package.json names index.js as main but no code files are present, and the skill declares no required environment variables or credentials even though integrating with those services would require API keys/tokens. This mismatch (claimed capabilities without code or declared connectors/credentials) is inconsistent and could mean the skill is just documentation or is incomplete.
Instruction Scope
Runtime instructions are high-level: install the skill, provide a YAML config (including local KB file paths), and start. The instructions only reference local KB files and standard openclaw commands — they do not tell the agent to read unrelated system files or to exfiltrate data. However they are vague about how to connect to external channels and how credentials will be provided/used, leaving broad discretion and potential privacy/behavior gaps.
Install Mechanism
No install spec and no code files beyond package.json — instruction-only skill. That is low install risk (nothing is downloaded or written by an installer). Note: package.json references an index.js that is not present; that suggests the package is incomplete or missing implementation.
!
Credentials
The skill claims integrations that normally require multiple credentials (Jira API token, Feishu/DingTalk/WeChat bot tokens, Telegram bot token, email SMTP credentials, etc.), yet requires.env is empty and SKILL.md does not list or name any environment variables. This omission is disproportionate and ambiguous: either the skill expects the platform to provide credentials implicitly, or the author omitted critical security details. That ambiguity increases risk because users might be prompted later to supply sensitive tokens without clear guidance.
Persistence & Privilege
The skill does not request always:true and is user-invocable with normal autonomous invocation allowed. There is no indication it modifies other skills or agent-wide configs. No persistence/privilege red flags in the provided manifest.
What to consider before installing
This bundle looks like a product brochure + config template rather than a complete implementation. Before installing or enabling it: 1) Ask the author for the missing source code (index.js/connector code) and for an explicit list of required credentials and environment variable names. 2) Do not provide broad/shared API keys — use least-privileged, service-specific tokens if you proceed. 3) Verify how personal or customer data is handled (where conversation logs and KB files are stored, retention, and external endpoints). 4) If connectors are provided later, review their code for hardcoded endpoints or data exfiltration, and prefer self-hosted or vetted integration components. 5) If you only want documentation, keep it disconnected from runtime; do not paste production credentials into the config until you audit the integration code.

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

Current versionv1.0.0
Download zip
latestvk97dqhm3e4aakbv2ngwrjkcd6x834k7z

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

SKILL.md

AI Customer Service Automation

7x24小时智能客服系统。

核心能力

✅ 智能问答:自动回复 80% 常见问题 ✅ 情感分析:识别愤怒/失望/满意,自动升级 ✅ 工单系统:复杂问题自动创建工单 ✅ 多渠道支持:飞书/微信/钉钉/Telegram/邮件 ✅ 数据看板:响应时间/解决率/满意度 ✅ 知识库同步:自动学习历史对话

快速开始

1. 安装

openclaw skills install ai-customer-service-automation

2. 配置

name: ai-customer-service-automation
config:
  # 知识库
  knowledgeBase:
    files:
      - ./docs/faq.md
      - ./docs/policy.pdf
      
  # 情感阈值
  sentiment:
    alertThreshold: 0.3  # 愤怒 < 0.3 自动升级
    
  # 工单系统
  ticket:
    provider: jira  # or feishu/dingtalk
    autoCreate: true
    
  # 多渠道
  channels:
    - feishu
    - wechat
    - telegram

3. 启动

openclaw start

ROI 计算

传统客服 vs AI 客服

指标传统客服AI 客服提升
响应时间5-10 分钟10 秒30-60x
人力成本¥5000/月/人¥500/月10x
工作时间8 小时24 小时3x
并发能力1 对 1无限
满意度70%85%+15%

定价

套餐价格功能
基础版¥199智能问答 + 1 渠道
专业版¥499情感分析 + 工单 + 3 渠道
企业版¥1499全功能 + 定制开发

客户案例

某电商平台

  • 日均咨询:5000+
  • AI 自动处理:4000+
  • 人工介入:20%
  • 客服成本:降低 60%

某 SaaS 公司

  • 用户数:10000+
  • AI 回答准确率:92%
  • 客户满意度:4.5/5
  • 客服团队:从 10 人降至 3 人

技术支持


安装配置服务:¥199 起,1 小时搞定!

Files

2 total
Select a file
Select a file to preview.

Comments

Loading comments…