Skill flagged — suspicious patterns detected

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

Resume-to-Tags

v1.0.0

从简历到纯标签矩阵的完整流程。接受简历文本/文件 → LLM 提取原子标签(含近义词扩展) → 创建飞书多维表格(多选标签字段) → 批量录入候选人 → 清理空白行列 → 输出可搜索的人才标签库。当用户提供简历并要求"生成标签矩阵"、"人才标签库"、"简历转表格"、"候选人打标"时使用。

0· 39·0 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 tuobadaidai/resume-to-tags.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Resume-to-Tags" (tuobadaidai/resume-to-tags) from ClawHub.
Skill page: https://clawhub.ai/tuobadaidai/resume-to-tags
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 tuobadaidai/resume-to-tags

ClawHub CLI

Package manager switcher

npx clawhub@latest install resume-to-tags
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
!
Purpose & Capability
The SKILL.md instructs the agent to create and modify Feishu Bitable resources (create app, patch table, create fields, batch insert records), which normally requires Feishu API credentials or a connector. The skill declares no required env vars or primary credential, so the declared requirements do not match the practical capability needed to perform Feishu operations.
!
Instruction Scope
The instructions and the included script explicitly extract phone numbers and emails and build a prompt containing resume text to be sent to an LLM. That means sensitive PII from resumes will be packaged into prompts/outputs and (per instructions) sent to an LLM and later inserted into an external table — there is no guidance about redaction, consent, or which LLM endpoint is safe to use. The instruction set otherwise stays within the stated purpose of tag extraction and table creation, but the PII handling and unspecified LLM/Feishu endpoints are significant scope/privacy concerns.
Install Mechanism
No install spec; the skill is instruction-only with a small helper script. Nothing is downloaded or installed automatically, so there is low install risk.
!
Credentials
There are zero required environment variables declared even though the workflow requires interacting with Feishu Bitable (and likely needs app tokens/credentials). The absence of declared credentials is disproportionate and creates ambiguity about where credentials come from (implicit agent connectors, or missing). Additionally, the script extracts PII (phone/email) that may be sent to an LLM or external service — sensitive environment/credential handling is not addressed.
Persistence & Privilege
always is false and the skill does not request permanent/global privileges or modify other skills' configurations. It does not autonomously persist credentials or enable itself. No elevated persistence concerns observed.
What to consider before installing
This skill generally does what it says (extract tags and populate a Feishu Bitable) but has two practical gaps you should deliberate before installing: (1) Feishu operations require API credentials or an agent connector — the skill does not declare or document which env vars/tokens are needed or how auth is handled. Confirm how your agent will authenticate to Feishu and restrict the credentials to least privilege. (2) The included script extracts phone numbers and emails and packages resume text into an LLM prompt; that will expose PII to whichever LLM or agent tool executes the prompt. If you plan to use real resumes, ensure you (a) have candidate consent, (b) redact or minimize PII before sending to third-party LLMs, or (c) use an on-prem/private model. Also verify the Feishu field IDs and destructive actions (deleting fields) against a test workspace first, and require explicit approval before batch-inserting real candidate data. If the author provides explicit documentation of how Feishu auth is sourced (specific env vars or a required connector) and adds privacy guidance (PII handling / recommended LLM endpoints), my confidence would increase and the skill would be closer to benign.

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

latestvk977d7hwhpcwc89ysk2q42pwvh85dvpp
39downloads
0stars
1versions
Updated 1d ago
v1.0.0
MIT-0

Resume-to-Tags — 简历到纯标签矩阵

从简历提取原子标签,构建可搜索、可匹配的人才标签矩阵。

核心设计理念

描述 → 标签:将简历中的描述性内容拆解为标准化原子标签,使人才匹配 = 标签重合度计算。

标签拆解规则

维度原始描述拆解标签
院校香港城市大学海外 硕士 QS前100 社科
院校同济大学本科 985 211 理工科
院校清华大学本科 C9 985 211
公司携程集团大厂 OTA 旅游 上市公司
公司京东集团大厂 电商 上市公司
公司明略科技AI创业
技能LLMOps/RAG链路优化LLMOps RAG LLM应用
技能大模型知识库RAG 知识库 LLM应用

完整工作流

Step 1: 标签提取

从简历文本提取原子标签,输出标准 JSON。

python3 skills/resume-to-tags/scripts/extract_tags.py --text "简历内容..."

或读取文件:

python3 skills/resume-to-tags/scripts/extract_tags.py --file resume.txt

Step 2: 近义词扩展

提取后自动扩充近义词标签,提高匹配覆盖率。近义词映射表见 references/synonyms.json

Step 3: 创建多维表格

使用飞书工具创建表格结构:

  1. 创建应用:feishu_bitable_app (action=create)
  2. 重命名默认表:feishu_bitable_app_table (action=patch)
  3. 重命名主字段为"姓名":feishu_bitable_app_table_field (action=update, fldFdaHuOL)
  4. 创建 7 个多选标签字段 + 1 个数字字段:
字段名类型说明
姓名文本(1)主键
院校标签多选(4)985/211/C9/QS前100/海外/本科/硕士...
公司标签多选(4)大厂/OTA/电商/AI创业/外企/旅游...
技能标签多选(4)RAG/Agent/招聘管理/数据分析...
工具标签多选(4)LangChain/Dify/SQL/SPSS...
领域标签多选(4)AI/LLM/HR/BD/产品/咨询...
语言标签多选(4)中文母语/英语工作/日语...
证书标签多选(4)初级经济师/CET6/N3...
工作年限数字(2)总经验年数
  1. 删除默认空白字段(单选 fld21nkFl6、日期 fld8rfpQsc、附件 fld84GszEh): feishu_bitable_app_table_field (action=delete)

Step 4: 录入候选人

批量插入记录:feishu_bitable_app_table_record (action=batch_create)

MultiSelect 字段的值会自动创建新选项(无需预定义)。

Step 5: 清理空白行列

删除默认空白列:

feishu_bitable_app_table_field: delete fld21nkFl6 (单选)
feishu_bitable_app_table_field: delete fld8rfpQsc (日期)
feishu_bitable_app_table_field: delete fld84GszEh (附件)

标签分类体系

院校标签

985 211 双一流 C9 G5 常春藤 QS前50 QS前100 QS前200 海外 本科 硕士 博士 理工科 社科 商科 在职教育

公司标签

大厂 互联网 OTA 旅游 电商 金融 AI创业 外企 国企 上市公司 独角兽 ToB ToC 本地生活 支付 物流 制造 咨询 教育 医疗 游戏 社交 内容 SaaS

技能标签(含近义词映射)

标准标签近义词/变体
RAG检索增强生成、大模型知识库、RAG链路优化
Agent智能体、Agentic Workflow、AI Agent
LLM应用大模型应用、LLM应用开发
微服务微服务架构、分布式架构
高并发高并发架构、高并发异步处理
PromptPrompt Engineering、提示词工程
数据分析数据清洗、回归分析、统计分析
人才寻源人才搜寻、Sourcing
胜任力模型胜任力画像构建、胜任力模型
招聘漏斗招聘漏斗分析
人才Mapping人才地图、Competitor Mapping
业务拓展BD、商务拓展
大客户管理KA管理、Key Account
OTA运营OTA渠道运营
市场趋势市场趋势分析
收益管理定价策略、Revenue Management
合同谈判商务谈判
客户关系客户关系维护、CRM
战略规划SP/BP、战略规划制定
知识萃取知识工程
抽象建模系统建模
产品架构产品架构设计
解决方案解决方案孵化
生态合作合作伙伴管理
AI中台AI平台建设、AI中台建设
人脸识别人脸技术
图像识别CV、计算机视觉
智能语音语音识别、TTS、ASR
大数据大数据架构
NLP自然语言处理

工具标签

LangChain Dify n8n OpenClaw Pinecone Milvus MySQL Redis Kafka ClickHouse Docker K8S AWS 阿里云 SQL SPSS Stata Pandas Excel CRM OTA平台 AI中台 OCR 猎聘 LinkedIn

领域标签

AI/LLM HR/招聘 BD/销售 产品 技术架构 OTA/旅游 电商 金融 数据分析 咨询 教育 医疗 游戏 社交

语言标签

中文母语 英语工作 英语流利 日语 CET6 托福 雅思

证书标签

初级经济师 PMP CET6 日语N3 日语N2 数学竞赛 英语专八

人才匹配逻辑

给定岗位需求标签集合 R,候选人标签集合 C:

  • 匹配度 = |R ∩ C| / |R| × 100%
  • 可按匹配度排序推荐候选人

相关文件

  • scripts/extract_tags.py — LLM 标签提取脚本
  • references/synonyms.json — 近义词映射表(标准标签 → 变体列表)
  • references/taxonomy.md — 完整标签分类体系

Comments

Loading comments...