Skill flagged — suspicious patterns detected

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

HR面试评价助手

v1.0.0

🎯 智能面试评估助手 通过口令触发,自动结合JD文字+简历文档+面试记录文档,生成专业面试评估报告 【触发口令】 面试评估、生成评估报告、候选人评估、面试评价、评估候选人 【使用方式】 1. 发送触发口令(如'面试评估') 2. 粘贴JD文字 3. 上传简历PDF/Word 4. 上传面试记录PDF/Word(...

1· 245·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for cloudmusiccio/hr-interview-evaluator.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "HR面试评价助手" (cloudmusiccio/hr-interview-evaluator) from ClawHub.
Skill page: https://clawhub.ai/cloudmusiccio/hr-interview-evaluator
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

Bare skill slug

openclaw skills install hr-interview-evaluator

ClawHub CLI

Package manager switcher

npx clawhub@latest install hr-interview-evaluator
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill's name/description (HR interview evaluator with resume/interview-record parsing and report export) aligns with requiring file parsing and PDF generation. However the metadata lists dependencies (file-parse, nos-cli, weasyprint) and the SKILL.md calls out local scripts (scripts/export_report.py, scripts/generate_radar.py) that are not present in the package. 'weasyprint' (PDF export) is reasonable; 'nos-cli' is unexplained and may be unrelated. Requiring tools or components that aren't provided is an incoherence.
!
Instruction Scope
The SKILL.md instructs the agent to request JD text and uploaded resume/interview files and to run parsing and export workflows (including Python scripts) and to call LLM-based parsing functions. It also instructs generating and exporting PDF/PNG files. There is no guidance on where uploaded candidate data is processed/stored or whether data is sent to external services. The instructions reference scripts and code (export/generate scripts) that are not included in the skill bundle, meaning the runtime behavior is underspecified and may rely on external tooling or other skills.
Install Mechanism
This is an instruction-only skill with no install spec or code files (low install risk). However, the declared requirements (weasyprint, nos-cli, file-parse) imply runtime dependencies that the skill does not install or provide; absent installation instructions, it's unclear whether the platform must supply these or if the skill expects arbitrary execution environment changes.
Credentials
The skill requests no environment variables, no credentials, and no config paths — appropriate for an on-platform document-processing helper. That said, the SKILL.md uses LLM parsing calls (parse_jd_with_llm) but does not disclose whether candidate data will be sent to external APIs or logged; the lack of declared endpoints/credentials leaves data flow unclear.
Persistence & Privilege
always is false and the skill is user-invocable (normal). The skill does include activation keyword/file triggers, but there is no indication it requests persistent system-wide privileges or modifies other skills. No evidence it writes permanent credentials or config.
Scan Findings in Context
[no_regex_findings] expected: Regex scanner found nothing to analyze because this is an instruction-only skill with no code files. That is expected, but absence of findings is not evidence that runtime behavior is safe.
What to consider before installing
Before installing, ask the publisher these questions: (1) Where do uploaded resumes and interview records get processed and stored? Do they leave your platform or the organization's network? (2) Provide the missing code or confirm which platform-provided components implement scripts/export_report.py and scripts/generate_radar.py referenced in SKILL.md. (3) Explain what 'nos-cli' is and why it's required; provide install instructions or justification for all declared dependencies. (4) Confirm data retention, logging, and deletion policies for candidate PII and whether generated reports are persisted or transmitted. (5) If you require local execution of export scripts, ensure the platform has trusted implementations (weasyprint, Python) and vet them. If these questions are not answered satisfactorily, do not enable the skill for sensitive candidate data.

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

latestvk97cehshkwr9kgfct6env8cfa18362ha
245downloads
1stars
1versions
Updated 22h ago
v1.0.0
MIT-0

HR智能面试评估助手

🎬 使用流程(三步完成)

第一步:发送触发口令

发送以下任意口令启动评估:

  • 面试评估
  • 生成评估报告
  • 候选人评估
  • 面试评价

第二步:提供JD信息

直接粘贴JD文字内容,例如:

岗位:前端开发工程师(P5)
要求:
1. 3-5年前端开发经验
2. 精通React或Vue框架
3. 熟悉TypeScript
4. 有大型项目经验
5. 薪资范围:20k-35k

第三步:上传文档

上传以下文档(支持PDF、Word格式):

  • 📄 简历文档(必需)- 候选人简历
  • 📝 面试记录(可选)- 面试官评价、反馈记录

📋 完整示例

用户操作流程

你: 面试评估

系统: 请输入岗位JD信息(直接粘贴文字即可):

你:

前端开发工程师招聘JD

岗位职责:
1. 负责公司核心产品前端开发
2. 参与前端架构设计
3. 优化前端性能

任职要求:
1. 3-5年前端经验
2. 精通React,熟悉Vue
3. 熟练使用TypeScript
4. 有大型项目经验优先

薪资:20k-35k

系统: 请上传候选人简历(PDF/Word):

你: 📎 张三简历.pdf

系统: 是否上传面试记录?(如无需上传请回复"跳过")

你: 📎 张三面试记录.docx

系统:

📊 正在分析...
✅ JD解析完成
✅ 简历解析完成  
✅ 面试记录解析完成
🎯 计算人岗匹配度...
📄 生成评估报告...

══════════════════════════════════════════════════
🎯 面试评估报告 - 张三
══════════════════════════════════════════════════

📋 岗位信息
岗位:前端开发工程师(P5)
薪资范围:20k-35k
要求年限:3-5年

👤 候选人信息
姓名:张三
学历:本科 / 浙江大学 / 计算机科学
工作年限:3年
当前职位:高级前端开发

══════════════════════════════════════════════════
📊 五维评估评分
══════════════════════════════════════════════════

┌──────────────┬────────┬────────────────────────┐
│ 评估维度     │ 评分   │ 说明                   │
├──────────────┼────────┼────────────────────────┤
│ 技术能力     │ ⭐⭐⭐⭐☆ │ React/TS精通,基础扎实  │
│ (25%)      │ 8.5/10 │                        │
├──────────────┼────────┼────────────────────────┤
│ 专业经验     │ ⭐⭐⭐⭐☆ │ 大型项目经验丰富        │
│ (20%)      │ 8.0/10 │ 电商行业匹配           │
├──────────────┼────────┼────────────────────────┤
│ 学习能力     │ ⭐⭐⭐⭐⭐ │ 主动学习新技术          │
│ (20%)      │ 9.0/10 │ 有技术分享经验         │
├──────────────┼────────┼────────────────────────┤
│ 岗位匹配     │ ⭐⭐⭐⭐☆ │ 技能与JD高度匹配        │
│ (20%)      │ 8.5/10 │ 经验年限符合           │
├──────────────┼────────┼────────────────────────┤
│ 稳定性       │ ⭐⭐⭐☆☆ │ 2年跳槽一次,可接受     │
│ (15%)      │ 6.0/10 │ 职业规划清晰           │
└──────────────┴────────┴────────────────────────┘

综合评分:8.1/10 ⭐⭐⭐⭐☆

══════════════════════════════════════════════════
🔴🟡🟢 风险评估
══════════════════════════════════════════════════

🟢 低风险:
   • 技能完全匹配JD要求,入职即可上手
   • 技术基础扎实,通过面试验证

🟡 中风险:
   • 期望薪资28k,处于JD范围上限
   • 管理经验较少,如需要带团队需培养

🔴 高风险:无

══════════════════════════════════════════════════
💡 智能录用建议
══════════════════════════════════════════════════

【建议类型】✅ 建议录用

【综合判断】
候选人综合评分8.1/10,属于优质候选人。
技术能力与JD要求高度匹配,项目经验符合期望,
面试表现优秀,建议尽快发放offer。

【匹配度分析】
✅ 必需技能100%匹配:React、TypeScript、Vue均满足
✅ 工作年限3年,符合JD要求
✅ 有大型电商项目经验,与业务匹配
✅ 期望薪资28k在预算范围内
⚠️ 管理经验欠缺(非JD必需项)

【录用条件】
• 薪资:28k(需确认预算审批)
• 级别:P5
• 试用期:3个月
• 入职时间:协商确定

【试用期考察重点】
1. 项目交付能力和代码质量
2. 团队协作和沟通表现
3. 业务理解和需求把控能力

══════════════════════════════════════════════════
📎 附件导出
══════════════════════════════════════════════════

已生成以下文件:
📄 张三_面试评估报告.pdf
🖼️ 张三_面试评估报告.png
📊 五维评分雷达图.png

══════════════════════════════════════════════════

🎯 技能激活逻辑

触发条件

当检测到以下任一情况时,自动激活本技能:

activation:
  # 口令触发
  keywords:
    - "面试评估"
    - "生成评估报告"
    - "候选人评估"
    - "面试评价"
    - "评估候选人"
    - "hiring recommendation"
    - "interview evaluation"
  
  # 文件上传触发(文件名含关键词)
  file_keywords:
    - "简历"
    - "面试"
    - "candidate"
    - "evaluation"
  
  # 文件类型
  file_types:
    - ".pdf"
    - ".doc"
    - ".docx"

对话流程

用户发送口令
    ↓
系统:请提供JD信息
    ↓
用户粘贴JD文字
    ↓
系统解析JD → 提取岗位要求
    ↓
系统:请上传简历
    ↓
用户上传简历.pdf
    ↓
系统解析简历 → 提取候选人信息
    ↓
系统:是否上传面试记录?
    ↓
用户:上传/跳过
    ↓
系统联合分析所有信息
    ↓
生成完整评估报告
    ↓
输出报告 + 导出PDF/PNG

📊 五维评分算法

评分维度与权重

维度权重评分依据满分
技术能力25%JD技能要求 vs 候选人掌握技能10分
专业经验20%年限、行业、项目规模匹配度10分
学习能力20%成长速度、技术广度、主动性10分
岗位匹配20%职责理解、文化契合、薪资匹配10分
稳定性15%跳槽频率、职业规划清晰度10分

评分标准

⭐⭐⭐⭐⭐ (9-10分) = 优秀 - 超出期望
⭐⭐⭐⭐☆ (7-8.9分) = 良好 - 符合期望
⭐⭐⭐☆☆ (5-6.9分) = 一般 - 基本满足
⭐⭐☆☆☆ (3-4.9分) = 较差 - 存在明显不足
⭐☆☆☆☆ (1-2.9分) = 差 - 不满足要求

综合得分计算

def calculate_overall_score(dimensions):
    """
    计算综合得分
    """
    weights = {
        'technical': 0.25,
        'professional': 0.20,
        'learning': 0.20,
        'role_fit': 0.20,
        'stability': 0.15
    }
    
    overall = sum(
        dimensions[dim] * weights[dim] 
        for dim in weights
    )
    
    return round(overall, 1)

💡 录用建议决策逻辑

决策矩阵

综合评分风险等级建议类型说明
≥ 8.0🟢 低✅ 建议录用优质候选人,尽快offer
7.0-7.9🟡 中⚠️ 条件录用基本合格,有条件通过
6.0-6.9🟡 中⚠️ 条件录用有短板,需谨慎评估
< 6.0🔴 高❌ 不建议录用不符合要求

条件录用的典型情况

当出现以下情况时,建议"⚠️ 条件录用":

  • 评分7.0-7.9,但存在可改进的短板
  • 个别技能不满足,但可以通过培训补足
  • 期望薪资略高,需要协商
  • 管理经验不足,但技术能力优秀
  • 试用期需要重点考察某些方面

📄 报告导出

支持格式

  1. PDF报告 - 适合:

    • 发送给用人部门审批
    • 归档保存
    • 打印纸质版
  2. PNG图片 - 适合:

    • 插入PPT汇报
    • 微信/钉钉分享
    • 邮件正文展示
  3. 雷达图 - 适合:

    • 直观展示五维能力
    • 对比多个候选人

导出命令

# 生成PDF
python3 scripts/export_report.py report.md 张三_评估报告 --format pdf

# 生成PNG
python3 scripts/export_report.py report.md 张三_评估报告 --format png

# 生成雷达图
python3 scripts/generate_radar.py scores.json 张三_雷达图.png

🔧 技术实现

核心流程代码

async def interview_evaluation_workflow(user_input):
    """
    面试评估主流程
    """
    
    # 阶段1:检测触发
    if not is_trigger_keyword(user_input):
        return None
    
    # 阶段2:收集JD信息
    jd_text = await request_jd_text()
    jd_info = parse_jd_with_llm(jd_text)
    
    # 阶段3:收集简历
    resume_file = await request_resume_file()
    resume_info = await parse_resume_document(resume_file)
    
    # 阶段4:收集面试记录(可选)
    interview_file = await request_interview_record()
    interview_info = None
    if interview_file:
        interview_info = await parse_interview_document(interview_file)
    
    # 阶段5:联合分析
    analysis = analyze_candidate(jd_info, resume_info, interview_info)
    
    # 阶段6:生成报告
    report = generate_evaluation_report(analysis)
    
    # 阶段7:导出文件
    pdf_path = export_to_pdf(report)
    png_path = export_to_png(report)
    
    return {
        'report': report,
        'files': [pdf_path, png_path]
    }

关键Prompt

JD解析Prompt

请从以下JD文字中提取结构化信息:

JD内容:
{jd_text}

提取要求:
1. 岗位名称和级别
2. 薪资范围
3. 必需技能清单
4. 加分项技能
5. 工作年限要求
6. 其他硬性要求

以JSON格式返回。

简历解析Prompt

请从以下简历内容提取候选人信息:

简历内容:
{resume_text}

提取要求:
1. 基本信息(姓名、学历、联系方式)
2. 工作年限
3. 工作经历(公司、职位、时间)
4. 技能清单
5. 项目经验
6. 教育背景

以JSON格式返回。

评估生成Prompt

基于以下信息,生成面试评估报告:

【JD要求】
{jd_info}

【候选人信息】
{resume_info}

【面试评价】
{interview_info}

要求输出:
1. 五维评分(每项1-10分)
2. 风险评估(高/中/低)
3. 录用建议(建议/条件/不建议)
4. 详细理由

⚠️ 注意事项

  1. 隐私保护:候选人信息仅用于评估,注意数据安全
  2. 人工复核:AI评估仅供参考,最终决策需人工判断
  3. JD准确性:提供的JD信息越详细,评估结果越准确
  4. 面试记录:如有面试记录,评估会更精准

🚀 快速开始

只需发送:面试评估

然后按照提示提供JD和简历,系统自动完成全部流程!

Comments

Loading comments...