Salary Research

薪酬调研助手 - 查询岗位薪酬数据并生成分析报告

Security Scan

The skill's requirements and instructions are consistent with a CLI-based salary-research tool that delegates work to a local 'salary-natural' binary; nothing in the package suggests hidden exfiltration or unrelated privileges.

Purpose & Capability

Name/description (salary research) align with the declared dependency on a 'salary-natural' CLI. The skill is instruction-only and delegates functionality to that binary, which is appropriate for a CLI wrapper.

Instruction Scope

SKILL.md describes using public recruiting APIs and web scraping and shows example CLI commands that call 'salary-natural' or 'salary-data'. The runtime script (run.sh) simply checks for the binary and forwards arguments. The instructions do not ask the agent to read unrelated files or secret env vars.

Install Mechanism

There is no install spec (instruction-only) and the included run.sh does not download or write arbitrary code. The script suggests an npm link command in an example path, which is a developer instruction rather than an installer downloading external artifacts.

Credentials

The skill declares no required environment variables or credentials. While SKILL.md mentions integrating with APIs or enterprise data, no undeclared secrets are requested by the skill itself. If you use enterprise-data features you may need to provide your own files or credentials outside the skill.

Persistence & Privilege

always is false and the skill does not request persistent/system-wide changes. It is user-invocable and can be invoked autonomously by keyword triggers (the platform default), which is expected behavior for a helper skill.

Assessment

This skill is a thin wrapper that expects a local 'salary-natural' CLI to do the real work. Before installing/use: (1) ensure you trust and inspect the 'salary-natural' tool (its source or npm package) because that binary will perform network requests and any scraping; (2) be cautious when uploading internal/company data — the skill's examples include uploading enterprise CSVs but the skill doesn't declare where those files are sent or stored; (3) web scraping may hit third-party sites and could violate their terms — confirm acceptable use; (4) the run.sh contains a developer-oriented npm link path (harmless) but indicates local development assumptions. If you want stronger assurance, run 'salary-natural' in a sandbox or review its code before granting the skill active use.

薪酬调研助手技能

技能概述

薪酬调研助手是一个 CLI 工具，可以通过调用招聘网站的公开 API 或爬取数据来查询薪酬信息，并生成详细的薪酬分析报告。

触发条件（自然语言）

当用户提到以下关键词时自动触发：

• 薪资相关：薪资、工资、薪酬、薪水、待遇、收入水平、薪资水平
• 询问价格：多少钱、能拿多少、收入多少
• 查询动作：查一下、查询、看看、帮我查

示例：

• "产品经理在北京能拿多少钱"
• "帮我查一下 Java 工程师的薪资"
• "上海本科产品经理工资水平"
• "深圳前端工程师待遇怎么样"
• "魔都算法工程师收入水平"（支持城市别名）

功能特性

• ✅ 自然语言理解，无需记忆命令格式
• ✅ 智能参数提取（岗位/城市/经验/学历）
• ✅ 支持城市别名（魔都→上海、鹏城→深圳等）
• ✅ 缺失参数自动提示
• ✅ 查询指定岗位和城市的薪酬数据
• ✅ 提供行业对比分析
• ✅ 显示城市间薪酬差异
• ✅ 支持多种输出格式（Markdown/JSON/Text/CSV）

使用方法

方式 1：自然语言查询（市场数据）⭐

salary-natural --text "产品经理在北京能拿多少钱"
salary-natural --text "帮我查一下 Java 工程师的薪资，3-5 年经验"

方式 2：企业数据管理（内部数据）

# 下载数据模板
salary-data template --output 薪酬数据模板.csv

# 上传企业数据
salary-data upload --file 薪酬数据.csv

# 查询企业数据
salary-data query --position "前端工程师" --city "北京"
salary-data query --all

# 生成薪酬报告
salary-data report --output 薪酬报告.md

参数说明

• --position / -p: 必需参数，输入岗位名称
• --city / -c: 必需参数，输入城市名称
• --experience / -e: 工作经验要求（如：3-5 年）
• --education: 学历要求（如：本科）
• --output: 输出格式 (json, markdown, text, csv)，默认为 markdown

输出内容（简洁版）

• 💰 薪资范围（最低 - 最高）
• 📊 平均薪资
• 🏢 行业对比（最高薪行业）
• 🌆 城市对比（最高薪城市）

技术实现

• 基于 Node.js 的 CLI 应用
• 自然语言处理模块自动解析用户输入
• 集成招聘网站 API 或网页爬虫
• 支持多种输出格式转换