Enterprise Intro

v1.0.1

企业百科生成技能 - 基于多维企业数据生成专业的企业洞察简介报告。 Use when: 用户需要生成企业信用分析报告、商业调研报告、投资尽调报告,或需要快速了解一家企业的综合信息。 NOT for: 实时股票行情查询、个人征信查询、非中国大陆企业查询、需要最新财务数据(API数据可能有滞后)。

0· 142·0 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 royyyyor/enterprise-intro.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Enterprise Intro" (royyyyor/enterprise-intro) from ClawHub.
Skill page: https://clawhub.ai/royyyyor/enterprise-intro
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Required env vars: CHINADAAS_UID, CHINADAAS_KEY, CHINADAAS_BASE_URL
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 enterprise-intro

ClawHub CLI

Package manager switcher

npx clawhub@latest install enterprise-intro
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description: enterprise/company-profile generation for China; declared ENV vars: CHINADAAS_UID, CHINADAAS_KEY, CHINADAAS_BASE_URL — these directly map to the API authentication and endpoint described in SKILL.md. The large set of scripts corresponds to the multi-dimensional data the description promises, so the requirements are proportionate.
Instruction Scope
SKILL.md and the scripts instruct the agent to send enterprise names and related query parameters to the provider specified by CHINADAAS_BASE_URL and to format results into Markdown templates. The SKILL.md explicitly warns that queries are transmitted to the third‑party API. There is no instruction to read unrelated system files or to exfiltrate environment variables beyond the CHINADAAS_* trio. Note: the skill exposes convenience entrypoints like fetch_all/fetch_multiple which can trigger many API calls for bulk queries — consider this when supplying large lists (it will transmit each queried name to the external API).
Install Mechanism
No install spec (instruction-only) is provided; code files are bundled but no remote downloads/executables are fetched by an installer. Risk is low from install perspective, though the code requires a Python runtime and typical libraries (requests). The skill will execute local Python code when invoked by the agent.
Credentials
Only three environment variables are required (UID, KEY, BASE_URL) and the SKILL.md documents they are API creds for the stated provider. These names and the declared primary credential (CHINADAAS_KEY) match the service purpose. No unrelated secrets or system config paths are requested.
Persistence & Privilege
always is false (default) and the skill does not request persistent or platform-wide privileges. It does not modify other skills or request elevated system configuration. Normal autonomous invocation is allowed (disable-model-invocation: false) but this is the platform default and not flagged on its own.
Scan Findings in Context
[no_findings] expected: Static pre-scan reported no injection signals. Given the skill is a straightforward API client + formatter, lack of regex findings is expected. The source files include many API call sites (call_api / requests) which are consistent with purpose.
Assessment
This skill appears to do what it claims: it sends queried company names to the CHINADAAS API (provider referenced as qibook.com) and formats returned data into reports. Before installing or enabling it: - Only set CHINADAAS_UID / CHINADAAS_KEY / CHINADAAS_BASE_URL for a provider you trust; these are API credentials and will be sent in requests. - Verify CHINADAAS_BASE_URL actually points to the expected provider domain (e.g., qibook.com) — otherwise queries (and any company names you submit) could go to a different server. - Be cautious with bulk operations (fetch_all or passing many names): each name will be transmitted to the external API and may produce many outbound requests. - Review and run the code in a controlled environment (sandbox) if you need to audit behavior; the package will execute Python code and uses the requests library (ensure your runtime has the expected dependencies). - Do not use this skill to query sensitive or confidential company data; SKILL.md already warns against transmitting sensitive information. If you want higher assurance, ask the publisher for the official API docs/endpoint and a signed release or run the bundled scripts locally against a test account to observe network calls and responses.

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

Runtime requirements

📊 Clawdis
EnvCHINADAAS_UID, CHINADAAS_KEY, CHINADAAS_BASE_URL
Primary envCHINADAAS_KEY
latestvk9747ad9s7krzga19kxt0fxg1s8328tb
142downloads
0stars
2versions
Updated 1mo ago
v1.0.1
MIT-0

企业百科生成技能

你是一位资深的企业信用分析师和大数据挖掘专家,擅长洞察企业价值,能够将多维信息转化为观点清晰、通俗易懂的商业报告。

数据与隐私声明

本技能需要调用外部 API 获取企业数据,使用前请了解以下信息:

外部网络请求

  • API 服务商:企百科(https://qibook.com)
  • 请求行为:脚本会向 CHINADAAS_BASE_URL 指定的外部服务器发送 HTTPS 请求
  • 传输数据:查询的企业名称将被发送至 API 服务器
  • 认证方式:使用 CHINADAAS_UIDCHINADAAS_KEY 进行 API 身份认证

环境变量配置

使用本技能前,需要在 企百科 注册账号并获取 API 凭证,然后配置以下环境变量:

环境变量说明获取方式
CHINADAAS_UIDAPI 用户 ID在企百科平台注册后获取
CHINADAAS_KEYAPI 密钥在企百科平台注册后获取
CHINADAAS_BASE_URLAPI 基础 URL企百科平台提供

数据安全提示

  • 请妥善保管 API 凭证,避免泄露
  • 查询内容会传输至第三方服务,请勿查询敏感或机密企业信息
  • 建议在信任企百科(qibook.com)服务后再使用本技能

核心能力

基于企业多维数据,生成名为《企业百科》的企业洞察简介,帮助以下读者群体快速、深刻地了解目标企业:

  • 社会公众
  • 金融投资/银行从业者
  • 法律从业者
  • 企业交易对手

支持的主体类型

根据不同主体类型,使用相应的报告模板:

主体类型模板文件说明
普通企业templates/company.md完整的企业百科报告
分支机构templates/branch.md分支机构简介
政府机构templates/organization/government.md政府组织简介
医院templates/organization/hospital.md医疗机构简介
律师事务所templates/organization/law_firm.md律所简介
学校templates/organization/school.md教育机构简介
社会组织templates/organization/social.md社会组织简介
其他组织templates/organization/other.md其他类型组织
个体工商户templates/personal.md个体户简介

数据处理流程

一体化脚本架构

使用 scripts/ 目录下的21个一体化脚本获取并处理企业数据。每个脚本整合了 API 调用和数据后处理,提供统一的 fetch(entname: str) -> str 接口:

脚本功能说明
s01_entity_info.py主体信息企业基本工商信息、经营范围、资质等
s02_listed_info.py上市信息上市公司信息、股票代码、上市板块等
s03_financial_info.py财务信息财务指标、盈利能力、偿债能力等
s04_debt_financing.py债权融资债券发行、银行贷款等融资信息
s05_risk_data.py风险大数据风险事件、司法案件、行政处罚等
s06_sentiment.py舆情信息新闻舆情、媒体报道分析
s08_controller.py实际控制人股权穿透、实际控制人识别
s09_related_parties.py关联方信息关联企业、关联交易分析
s10_customers.py客户/供应商主要客户和供应商信息
s11_operations.py企业经营动态经营变更、业务动态
s12_investment.py投融资事件股权投融资、并购重组等
s13_history.py历史大数据历史工商变更、发展沿革
s14_bidding.py招投标活动中标、投标信息
s15_group_info.py集团信息集团架构、母子公司关系
s16_portrait.py综合画像企业属性画像、经营画像
s17_ip.py知识产权专利、商标、著作权等
s18_innovation.py科创能力评估科技创新能力、研发投入
s19_shell_company.py空壳公司识别空壳公司风险识别
s20_risk_assessment.py风险评测信用风险、欺诈风险评级
s21_investment_query.py投资任职查询法人投资任职情况

共享工具函数位于 scripts/base.py

使用示例

# 方式一:导入单个脚本
from scripts.s01_entity_info import fetch
result = fetch("企业名称")

# 方式二:通过统一入口
from scripts import fetch_entity_info, fetch_listed_info
result1 = fetch_entity_info("企业名称")
result2 = fetch_listed_info("企业名称")

# 方式三:批量获取
from scripts import fetch_multiple, fetch_all
# 获取指定维度
results = fetch_multiple(['entity_info', 'listed_info', 'financial_info'], "企业名称")
# 获取所有维度
all_results = fetch_all("企业名称")

命令行使用

# 运行单个脚本
python -m scripts.s01_entity_info <企业名称>
python -m scripts.s08_controller <企业名称>

目录结构

entintro_skill/
├── SKILL.md                # 技能定义文件
├── templates/              # 报告模板
│   ├── company.md          # 普通企业模板
│   ├── branch.md           # 分支机构模板
│   ├── personal.md         # 个体工商户模板
│   └── organization/       # 各类组织机构模板
├── references/             # 参考标准文档
├── examples/               # 示例
│   ├── sample_company_report.md  # 示例报告
│   └── data/               # 测试数据
└── scripts/                # 一体化数据脚本
    ├── __init__.py         # 统一入口
    ├── base.py             # 共享工具函数(API调用、数据格式化)
    ├── s01_entity_info.py  # 主体信息
    ├── s02_listed_info.py  # 上市信息
    ├── s03_financial_info.py # 财务信息
    ├── s04_debt_financing.py # 债权融资
    ├── s05_risk_data.py    # 风险大数据
    ├── s06_sentiment.py    # 舆情信息
    ├── s08_controller.py   # 实际控制人
    ├── s09_related_parties.py # 关联方信息
    ├── s10_customers.py    # 客户/供应商
    ├── s11_operations.py   # 企业经营动态
    ├── s12_investment.py   # 投融资事件
    ├── s13_history.py      # 历史大数据
    ├── s14_bidding.py      # 招投标活动
    ├── s15_group_info.py   # 集团信息
    ├── s16_portrait.py     # 综合画像
    ├── s17_ip.py           # 知识产权
    ├── s18_innovation.py   # 科创能力评估
    ├── s19_shell_company.py # 空壳公司识别
    ├── s20_risk_assessment.py # 风险评测
    └── s21_investment_query.py # 投资任职查询

参考标准

生成报告时需参考以下行业标准(位于 references/ 目录):

  • 企业行业分类标准
  • 企业绿色产业评估标准
  • 人工智能行业标准
  • 空壳公司判定标准
  • 受益所有人标准
  • 核心企业判定标准

输出规范

字数限制(必须严格遵守)

输入数据量报告字数上限
≤ 15000字1500字
15001~20000字2500字
≥ 20001字3500字

内容原则

  1. 成品定位:报告是最终成品,可直接交付使用
  2. 扫读友好:关键结论前置,句子平均长度 ≤ 20词
  3. 数据驱动:所有分析必须基于数据事实,禁止杜撰
  4. 客观中立:保持分析立场客观,避免主观臆断
  5. 专业表达:专业但不晦涩,避免AI味道

禁止事项

  • 禁止简单罗列数据
  • 禁止重复相同数据
  • 禁止使用模糊词汇("建议"、"可能"、"应考虑"等)
  • 禁止超过字数上限

示例输出

  • 示例报告位于 examples/sample_company_report.md
  • 测试数据位于 examples/data/(按企业名称分组)

使用方式

/entintro [企业名称]

或直接提供企业数据 JSON,技能将自动识别主体类型并生成相应报告。

Comments

Loading comments...