Three-Statement Model Builder
ginlix-ai/LangAlpha
Builds or repairs an integrated income statement, balance sheet and cash flow model in Excel with live formulas, supporting schedules, scenarios and a Checks sheet.
Looks up Chinese company registration data in the CICPA industry knowledge base, from a quick search to a full 61-dimension export, subsidiary discovery and Excel output.
SKILL.md written in Chinese; this summary is our English description.
Install CICPA Company Registration Query
$ npx skills add nigo81/nigo-skills --skill cicpa-company-query -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install nigo81/nigo-skills cicpa-company-query --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/nigo81/nigo-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/cicpa-company-query .claude/skills/cicpa-company-query && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "cicpa-company-query" agent skill from https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-query into .claude/skills/cicpa-company-query/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cicpa-company-query", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-queryType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add nigo81/nigo-skills --skill cicpa-company-query -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install nigo81/nigo-skills cicpa-company-query --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nigo81/nigo-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/cicpa-company-query .agents/skills/cicpa-company-query && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "cicpa-company-query" agent skill from https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-query into .agents/skills/cicpa-company-query/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cicpa-company-query", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add nigo81/nigo-skills --skill cicpa-company-query -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install nigo81/nigo-skills cicpa-company-query --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nigo81/nigo-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/cicpa-company-query .cursor/skills/cicpa-company-query && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "cicpa-company-query" agent skill from https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-query into .cursor/skills/cicpa-company-query/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cicpa-company-query", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/nigo81/nigo-skills.git --path cicpa-company-query--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add nigo81/nigo-skills --skill cicpa-company-query -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install nigo81/nigo-skills cicpa-company-query --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nigo81/nigo-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/cicpa-company-query .gemini/skills/cicpa-company-query && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "cicpa-company-query" agent skill from https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-query into .gemini/skills/cicpa-company-query/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cicpa-company-query", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install nigo81/nigo-skills cicpa-company-queryInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add nigo81/nigo-skills --skill cicpa-company-query -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/nigo81/nigo-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/cicpa-company-query .github/skills/cicpa-company-query && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "cicpa-company-query" agent skill from https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-query into .github/skills/cicpa-company-query/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cicpa-company-query", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add nigo81/nigo-skills --skill cicpa-company-query -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install nigo81/nigo-skills cicpa-company-query --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/nigo81/nigo-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/cicpa-company-query .opencode/skills/cicpa-company-query && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "cicpa-company-query" agent skill from https://github.com/nigo81/nigo-skills/tree/main/cicpa-company-query into .opencode/skills/cicpa-company-query/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cicpa-company-query", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
cicpa-company-query- GitHub stars
- 133
- Token cost
- ~2.8k tokens
- SKILL.md length
- 583 words
- Files
- 3 (incl. scripts)
- Skills in repo
- 13
- Repo updated
- First seen
- Licence
- MIT
At a glance
Looks up Chinese company registration data in the CICPA industry knowledge base, from a quick search to a full 61-dimension export, subsidiary discovery and Excel output.
- Works in 2 steps: 依赖：pip install requests openpyxl（浏览器登录还需… → 无需配置文件
- Checking a company's registered capital, legal representative or registered address
- SKILL.md covers Overview, Mode Selection ⚠️ 关键决策, Quick Reference and When to Use, plus 5 more sections
- Runs Python scripts from its folder; calls python3 and pip
What it does
The skill queries business registration information through the CICPA industry knowledge base at zsk-cmis.cicpa.org.cn, the research system of the Chinese Institute of Certified Public Accountants. It has four modes: a light single-company search, a detail view with shareholders, key personnel and equity structure, a basic batch query of 22 fields into one Excel file, and a full export of 61 dimensions packed as a ZIP of 53 Excel files. A separate option discovers subsidiaries, with a default shareholding threshold of 50 percent.
It runs only through scripts/cicpa_query.py and never through the MCP tool of the same name, which the skill says crashes on a known Playwright sync and async bug. The agent logs in through a controlled browser, saves cookies to .cicpa_cookies.json in the current working directory and reuses them for 24 hours, so commands must run from the project directory. When you ask for the full export it goes straight there instead of running a basic query first. The SKILL.md is in Chinese, and other skills can import its functions.
When your agent uses it
- Checking a company's registered capital, legal representative or registered address
- Pulling a full company profile for due diligence or audit working papers
- Finding subsidiaries of a company by shareholding
- Batch-querying a list of companies into an Excel sheet
Example prompts
- “Look up the registered address and registered capital of Alibaba (China) Co., Ltd.”
- “Run a full export of all dimensions for this company for our due diligence file.”
- “Discover the subsidiaries of this group where it holds at least 50 percent.”
Requirements
- Python to run scripts/cicpa_query.py
- A browser login to the CICPA knowledge base to obtain cookies
Workflow steps
2 steps, taken from the first numbered list in SKILL.md.
- 依赖：pip install requests openpyxl（浏览器登录还需 playwright）
- 无需配置文件
What it can do on your machine
Read from SKILL.md and the folder at commit 6468211. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 1 file in scripts/ (Python), which the agent can run.
Shell commands in SKILL.md call:
python3pip
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use pip, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
CICPA Company Registration Query loads about 2.8k tokens when it runs. Until then it costs about 99 tokens; SKILL.md has 583 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from nigo81/nigo-skills at commit 6468211, republished under its MIT licence (© nigo81). 583 words, ~2,841 tokens.
Download SKILL.mdSave it as
.claude/skills/cicpa-company-query/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.- name
- cicpa-company-query
- description
- Use when querying company business registration info (工商信息) from CICPA system (中注协行业知识库). Triggers on 工商信息, 企业查询, 法人, 注册资本, 注册地址, 地址核查, 查公司, 企业详情, 子公司, 关联方识别的子公司发现, 单个公司查询. Full export (--browser-export) triggers on 全量, 完整维度, 所有维度, 60+维度, 60多个维度, 53个Excel, 全部工商信息, 完整企业画像, 尽调, 尽职调查. Also triggers on 批量查企业, 注协查询, cicpa query, audit working papers needing company background.
- slug
- nigo-cicpa-company-query
- displayName
- 中注协企业工商查询
- version
- 1.0.0
- summary
- 从中注协行业知识库查询企业工商注册信息（法人、注册资本、注册地址等），支持单个查询和全量60+维度导出。
- license
- MIT
注协系统工商信息查询
⛔ 绝对禁止使用 MCP 工具
sw-audit-toolbox_cicpa_query的 export/subsidiary 模式有已知的 Playwright sync/async 冲突 bug， 在当前环境调用必定崩溃。必须且只能使用 CLI 脚本。所有命令见下方 Quick Reference。
如果你在工具列表中看到了 MCP 工具，忽略它们，按照下方 CLI 命令执行。
Overview
通过中注协行业知识库（zsk-cmis.cicpa.org.cn）查询企业工商信息。四种模式：
- 轻量搜索：单企业关键词搜索，返回基础信息 + org_id
- 企业详情：单企业完整详情（工商信息 + 股东 + 主要人员 + 股权结构）
- 基础查询：批量查询 22 个字段，输出单 Excel
- 完整导出：61 个维度、53 个 Excel 文件打包 ZIP（纯 API）
无需配置文件，通过 AI 控制浏览器登录获取 cookies，24 小时内复用。
⚠️ Cookie 文件位置（重要）
Cookie 文件保存在当前工作目录（Path.cwd()）下的 .cicpa_cookies.json，不是用户主目录。
AI 执行命令时必须确保工作目录正确：
- 如果在项目目录下执行，cookie 就在项目目录下
related-party-identification等其他 skill 调用时，也在同一工作目录找 cookie- 不要在
~（主目录）下运行 cookie 相关命令，否则会保存到~/.cicpa_cookies.json，其他项目找不到
Mode Selection ⚠️ 关键决策
| 用户说的 | 模式 | 命令 |
|---|---|---|
| "查一下XX公司"、"XX的工商信息" | 轻量搜索 | search_company("XX") |
| "XX公司详情"、"XX股东、高管" | 企业详情 | --detail "XX" |
| "全量工商信息"、"完整维度"、"60多个维度" | 完整导出 | --browser-export |
| "查一下地址"、"注册资本多少" | 基础查询 | -n "企业名" |
| "批量查企业工商信息" | 基础查询 + --all-fields | -n ... --all-fields |
| "发现XX的子公司" | 子公司发现 | --discover-subsidiaries "XX" |
| 不确定 | 问用户 |
输出对比
| 轻量搜索 | 企业详情 | 基础查询 | 完整导出 | |
|---|---|---|---|---|
| 耗时 | <1 秒 | 2~3 秒 | 5~10 秒 | 30 秒~3 分钟 |
| 字段 | 基础 20+ | 工商+股东+人员+股权 | 10~22 个 | 61 维度 |
| 输出 | 返回 dict | 终端+JSON | 单 Excel | ZIP (53 个 Excel) |
| API | home_search (轻量) | 4 个详情 API | batch upload | batch export |
⚠️ 禁止：先用基础查询再补完整导出。 用户要"全量"就一步到位 --browser-export。
Quick Reference
SCRIPT=~/.claude/skills/cicpa-company-query/scripts/cicpa_query.py
# ===== 轻量搜索（Python 函数，非 CLI）=====
# 返回 [{name, org_id, legal_person, address, ...}]
# from cicpa_query import search_company
# results = search_company("华为技术有限公司")
# ===== 企业详情（--detail）=====
# 按名称搜索（自动获取 org_id）
python3 $SCRIPT --detail "华为技术有限公司"
# 按 org_id 直接查询
python3 $SCRIPT --detail "T003573795" -o output.json
# ===== 子公司发现（--discover-subsidiaries）=====
# 发现持股 >= 50% 的子公司
python3 $SCRIPT --discover-subsidiaries "天津卓朗科技发展有限公司"
# 自定义阈值
python3 $SCRIPT --discover-subsidiaries "天津卓朗科技发展有限公司" --subsidiary-threshold 30
# ===== 基础查询 =====
python3 $SCRIPT -n "华为技术有限公司"
# 基础查询（全部 22 个字段）
python3 $SCRIPT -n "企业名" --all-fields
# 从文件读取企业名单
python3 $SCRIPT -f companies.xlsx
# ===== 完整导出（53 个 Excel → ZIP）=====
python3 $SCRIPT -n "企业A" "企业B" --browser-export
# ===== Cookies 管理 =====
# ⚠️ Cookie 文件在当前工作目录下：./cicpa_cookies.json
# 确保用 workdir 参数指定正确目录，不要在 ~ 下运行
python3 $SCRIPT --check-cookies
python3 $SCRIPT --save-cookies '{"XSRF-TOKEN":"xxx","cicpa_token":"xxx",...}'
python3 $SCRIPT --login输出选项
| 参数 | 说明 |
|---|---|
--detail ID_OR_NAME | 单企业详情（org_id 或名称） |
--discover-subsidiaries NAME | 发现子公司（默认持股>=50%） |
--subsidiary-threshold N | 子公司持股阈值（配合上一参数，默认 50） |
-o PATH | 指定输出路径 |
--all-fields | 导出全部 22 个字段（默认 10 个） |
--by-dimension | 按维度分类导出多个 Excel → ZIP |
--browser-export | 完整导出（61 维度，53 个 Excel → ZIP） |
--save-cookies JSON | 保存 AI 抓取的 cookies（JSON 字符串） |
--login | 终端手动浏览器登录 |
Python API（供其他 skill 调用）
import sys
sys.path.insert(0, '<cicpa-scripts-dir>')
from cicpa_query import search_company, discover_subsidiaries, get_company_detail三个函数的区别（AI 选择依据）：
search_company | discover_subsidiaries | get_company_detail | |
|---|---|---|---|
| API 调用 | 1 次（home_search） | 2 次（search + equity） | 4 次（basic+holders+persons+equity） |
| 输入 | 关键词 | 企业名称 + 阈值 | org_id 或名称 |
| 返回 | [{name, org_id, ...}] | [{name, ratio, org_id}] | {basic_info, shareholders, ...} |
| 耗时 | <1 秒 | 1~2 秒 | 2~3 秒 |
| 何时用 | 只需 org_id 或确认企业存在 | 查"这家公司投资/控股了谁" | 查"这家公司的完整信息" |
# ① 只需 org_id 或基础信息 → search_company
# 场景：确认企业名称、拿 org_id 给其他函数用
results = search_company("华为技术有限公司")
# → [{"name": "华为技术有限公司", "org_id": "T003573795", "legal_person": "任正非", ...}]
# ② 查子公司/对外投资 → discover_subsidiaries
# 场景：关联方预查、合并下载名单
subs = discover_subsidiaries("审计目标", threshold=50)
# → [{"name": "子公司A", "ratio": 100.0, "org_id": "Txxx"}, ...]
# ③ 查一家公司的完整信息 → get_company_detail
# 场景：审计中了解交易对手、函证地址核查、单公司尽调
detail = get_company_detail("T003573795", "天津卓朗科技发展有限公司")
# → {
# "basic_info": {name, address, capital, legal_person, ...},
# "shareholders": {list: [{holder_name, held_ratio, ...}]},
# "main_persons": {tab_list, ...},
# "equity": {holders: {...}, invests: {...}}
# }完整导出包含的维度（5 大类 53 个文件）
| 维度 | 文件数 | 内容示例 |
|---|---|---|
| 基本信息 | 14 | 工商信息、股东、变更记录、分支机构、主要人员 |
| 经营情况 | 18 | 招投标、商标、专利、客户、供应商、税务信息 |
| 企业发展 | 7 | 对外投资、融资历史、核心团队、债券信息 |
| 经营风险 | 13 | 行政处罚、欠税、经营异常、股权出质、注销 |
| 司法风险 | 9 | 裁判文书、立案、开庭公告、被执行人、失信 |
When to Use
AI 决策指南：用户要什么 → 用哪个
| 用户说/需要 | 用什么 | 为什么 |
|---|---|---|
| "查一下这家公司"、"确认企业名称" | search_company() | <1 秒，只拿基础信息 |
| "这家公司投资了谁"、"查子公司" | --discover-subsidiaries | equity API 拿对外投资 |
| "这家公司详情"、"查股东高管" | --detail | 4 个 API 拿完整信息 |
| "批量查企业"、"查几个字段" | -n + --all-fields | batch 查询，输出 Excel |
| "全量"、"完整维度"、"60多个维度" | --browser-export | 61 维度 53 个 Excel |
特殊场景：被其他 skill 调用
related-party-identification预查子公司 → 调用discover_subsidiaries()related-party-identification需要拿 org_id → 调用search_company()- 函证地址核查只需地址/电话 →
search_company()已足够，不需要--detail
When NOT to Use
- 只需要天眼查/企查查的单个企业信息（用网页直接查更快）
- 需要 A 股上市公司财报数据（用 a-stock-financial skill）
- 需要法律法规查询（用 china-law-search skill）
前置条件
- 依赖：
pip install requests openpyxl（浏览器登录还需playwright） - 无需配置文件
Cookie 自动检查与获取流程
AI 在任何需要调用注协 API 的操作前，必须先检查 cookie。全自动化，不要问用户。
⚠️ 浏览器工具选择（强制）
| 工具 | 是否可用 | 说明 |
|---|---|---|
playwright-mcp_* (Playwright MCP) | ✅ 必须用这个 | 打开可见浏览器窗口，用户能看到并操作 |
agent-browser | ❌ 禁止使用 | 无头模式，用户看不到浏览器窗口，无法手动登录 |
脚本内 --login | ❌ 禁止使用 | 检测到非交互式终端会失败 |
原因：登录需要用户手动操作（拖滑块验证），必须打开用户可见的浏览器窗口。只有 playwright-mcp_* 系列工具能做到这一点。
Show full SKILL.md (228 more words)Show less
自动流程
① 检查本地 cookie（每次执行查询前自动做）
bash: python3 $SCRIPT --check-cookies
（⚠️ 用 workdir 指定项目目录）
② ✅ 有效 → 直接执行查询命令，无需任何登录操作
③ ❌ 无效/不存在 → AI 自动启动浏览器登录：
告诉用户："Cookie 已过期/不存在，正在打开浏览器，请手动登录"
步骤 1：打开登录页
调用 playwright-mcp_browser_navigate → https://cmis.cicpa.org.cn/#/login
→ 告诉用户："请在浏览器中登录（选用户类型、输入密码、拖滑块）"
步骤 2：用户说"已登录"后，点击「行业执业知识库」
调用 playwright-mcp_browser_snapshot → 找到"行业执业知识库"菜单项
调用 playwright-mcp_browser_click → 点击该菜单项
调用 playwright-mcp_browser_wait_for → 等待新标签页加载
步骤 3：导航到企业数据库并抓取 cookies
调用 playwright-mcp_browser_tabs → 切换到新打开的 zsk 标签页
调用 playwright-mcp_browser_navigate → https://zsk-cmis.cicpa.org.cn/companylibrarynew/
调用 playwright-mcp_browser_run_code_unsafe → 执行以下代码获取并转换 cookie：
```javascript
const cookies = await page.context().cookies();
const cookieDict = cookies.reduce((acc, c) => { acc[c.name] = c.value; return acc; }, {});
JSON.stringify(cookieDict);
```
⚠️ **Cookie 格式转换（必须）**: page.context().cookies() 返回 `[{name, value, ...}]` 数组，
上面的 JS 代码已经转换为 `{name: value}` 字典格式
步骤 4：保存并验证
bash（⚠️ workdir 必须是项目目录）:
python3 $SCRIPT --save-cookies '<步骤3返回的JSON>'
python3 $SCRIPT --check-cookies → 确认 ✅ 有效
④ 如果 Playwright MCP 工具不可用 → 退化为手动方式：
提示用户手动登录后粘贴 cookie，或在交互终端运行 python3 cicpa_query.py --loginPlaywright MCP 工具映射：
| 步骤 | 工具 | 说明 |
|---|---|---|
| 打开页面 | playwright-mcp_browser_navigate | 导航到 URL |
| 查看页面 | playwright-mcp_browser_snapshot | 获取页面快照（无障碍树） |
| 点击元素 | playwright-mcp_browser_click | 点击按钮/链接 |
| 切换标签 | playwright-mcp_browser_tabs | list/new/close/select |
| 等待加载 | playwright-mcp_browser_wait_for | 等待文本/元素/URL 变化 |
| 获取 cookies | playwright-mcp_browser_run_code_unsafe | 执行 page.context().cookies() |
| 截图调试 | playwright-mcp_browser_take_screenshot | 查看当前页面状态 |
备用方式：手动从浏览器 F12 抓取 cookies
仅在 Playwright MCP 不可用时使用：
1. 打开浏览器访问 https://cmis.cicpa.org.cn，手动登录
2. 登录成功后访问 https://zsk-cmis.cicpa.org.cn/companylibrarynew/
3. F12 → Network → 刷新 → 找「companylibrarynew」请求 → 复制 Cookie 请求头
4. 粘贴给 AI用户粘贴后，AI 解析 key1=value1; key2=value2 为 {key1: value1, key2: value2} 格式，然后保存：
python3 $SCRIPT --save-cookies '{解析后的JSON}'
python3 $SCRIPT --check-cookies为什么不用
document.cookie？document.cookie无法获取 HttpOnly 的 cookie（如__snaker__id、gdxidpyhxdE），只能拿到部分 cookie。
⚠️ 关键：必须走完整 SSO 流程
| 场景 | 获得的 cookies | 能否查询 |
|---|---|---|
| 只登录 cmis，不进知识库 | 4 个（不完整） | ❌ 报"未登陆用户" |
| 登录 cmis → 点击知识库 → 跳转 zsk | 9 个（完整） | ✅ 正常 |
必须的 9 个 cookies：XSRF-TOKEN、yuqing_whole_jsessionid、cicpa_token、cicpa_ticket、companyVerifyCode、userid、u_name、__snaker__id、gdxidpyhxdE
如果 --check-cookies 显示少于 9 个，说明 SSO 流程不完整，需要重新登录。
Cookie 文件格式说明
.cicpa_cookies.json 必须是以下格式的 JSON（保存到当前工作目录）：
{
"cookies": {
"XSRF-TOKEN": "xxx",
"yuqing_whole_jsessionid": "xxx",
"cicpa_token": "xxx",
"cicpa_ticket": "xxx",
"companyVerifyCode": "xxx",
"userid": "xxx",
"u_name": "xxx",
"__snaker__id": "xxx",
"gdxidpyhxdE": "xxx"
},
"session_token": "xxx", // 可选，从 localStorage 获取
"saved_at": 1719123456.789 // ⚠️ 必须字段，Unix 时间戳，用于检查 24 小时有效期
}关键点：
cookies字段必须是{name: value}格式，不是[{name, value}]数组saved_at字段是必需的，用于检查 cookie 是否超过 24 小时过期- 通过 Playwright 的
page.context().cookies()获取时需要转换格式
MCP 模式的已知限制
如果通过 MCP 工具（如 cicpa_query MCP）调用 export 或 subsidiary 模式：
- Cookie 失效时会触发浏览器登录：在无头模式下打开浏览器会导致崩溃
- 建议优先使用 CLI 模式：通过
--login参数在交互式终端中登录，或使用--save-cookies手动保存
注意点
python3 $SCRIPT --check-cookies
# → 显示 cookie 数量、名称、剩余有效期
# → ✅ 有效 或 ❌ 需要重新登录| 注意点 | 说明 |
|---|---|
| 企业名称必须精确 | 用工商注册全称（如"深圳市腾讯计算机系统有限公司"），简称无法匹配 |
| Cookies 24h 过期 | 过期后需重新登录 |
| 最多 10,000 家 | 单次批量查询上限 |
| 导出耗时 | 10 家约 45 秒，100+ 家可能 2-3 分钟 |
| 自动验证 | 完整导出后自动对比输入 vs 输出公司名称，报告匹配情况 |
Common Mistakes
| 问题 | 原因 | 解决 |
|---|---|---|
| 用户要"全量"却只给了基础查询 | 没看 Mode Selection，用了默认模式 | 用户说"全量/完整/所有维度"→ 必须用 --browser-export |
| 先用基础查询再补完整导出 | 分两步浪费时间 | 用户要全量就一步到位，不要多此一举 |
| Cookie 保存到了主目录 | 在 ~ 下执行命令，cookie 保存到 ~/.cicpa_cookies.json | 所有命令必须在项目工作目录下执行（bash 用 workdir 参数） |
| 其他 skill 找不到 cookie | 工作目录不一致，cookie 在另一个目录 | 确保 --save-cookies、--check-cookies、扫描脚本都在同一 workdir 下运行 |
| 查询报"未登陆用户" | 只有 4 个不完整 cookie | 重新登录，确保点了「行业执业知识库」，拿到 9 个 cookie |
| 上传后匹配 0 家 | 企业名称非注册全称 | 用精确的工商注册名 |
| Cookies 无效 | 超过 24 小时 | 重新登录获取 cookies |
| 导出轮询超时 | 数据量大或系统繁忙 | 等待后手动去下载中心下载 |
| XSRF 报错 | cookies 中缺少 XSRF-TOKEN | 重新登录 |
目录结构
~/.claude/skills/cicpa-company-query/
SKILL.md # 本文档
scripts/
cicpa_query.py # 主脚本作者：nigo 微信公众号：逆行的狗
© nigo81, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 2 other files (scripts) in cicpa-company-query of nigo81/nigo-skills.
- SKILL.md
- .gitignore
- scripts/cicpa_query.py
Open the folder on GitHubat commit 6468211
Compare with similar skills
CICPA Company Registration Query next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| CICPA Company Registration Query this skillnigo81/nigo-skills | 133 | — | ~2.8k | Automated safety check: Pass | MIT | |
| Three-Statement Model Builderginlix-ai/LangAlpha | 1.8k | — | ~5.4k | Automated safety check: Pass | Apache-2.0 | |
| Dcf ModelWind-Alice/AliceMarket | 134 | 2 repos | ~12k | Automated safety check: Pass | None | |
| Dcf ModelLuciole-Studio/Misaka-Agent | 171 | 2 repos | ~12k | Automated safety check: Pass | Apache-2.0 | |
| Datapack Builderw95/awesome-claude-corporate-skills | 244 | 1 repos | ~6k | Automated safety check: Pass | MIT | |
| Fsi Comps Analysisw95/awesome-claude-corporate-skills | 244 | 1 repos | ~6.7k | Automated safety check: Pass | MIT |
Similar skills
Dcf Model
Wind-Alice/AliceMarket
Real DCF (Discounted Cash Flow) model creation for equity valuation.
Dcf Model
Luciole-Studio/Misaka-Agent
Build discounted cash flow valuation workbooks in Excel. An agent skill from Luciole-Studio/Misaka-Agent.
Datapack Builder
w95/awesome-claude-corporate-skills
Build professional financial services data packs from various sources including CIMs, offering memorandums, SEC filings, web search, or MCP servers.
Fsi Comps Analysis
w95/awesome-claude-corporate-skills
Build institutional-grade comparable company analyses with operating metrics, valuation multiples, and statistical benchmarking in Excel/spreadsheet format.
Anti Gambling Trader
mars-tw/anti-gambling-trader-tw
分析台股 / 台股ETF / 台指期選擇權 / 美股 / 加密貨幣 / 外匯的交易紀錄 (CSV / JSON / Excel),用統計學判斷使用者的獲利是「可重複的優勢」還是 「運氣 + 倖存者偏差(賭博)」,不適合長期投資會明確勸退。內建反詐工具: 掃描群組對話的詐騙話術(scan-text)、檢驗老師宣稱的績效(guru-check)、…
More from nigo81/nigo-skills
All 13 skills in this repoAudit Report Checker
nigo81/nigo-skills
Checks audit reports (financial statements and notes) for cross-reference, summation and text errors, with arithmetic done by script and results exported to Excel and Markdown.
Bank Statement Merge and Reconciliation
nigo81/nigo-skills
Merges bank statements from several banks and formats into one standard Excel layout, then reconciles them against the general ledger with a layered matching engine.
Accounting Standards Advisor (Chen Yiwei)
nigo81/nigo-skills
A Chinese-language persona that answers accounting-standards and audit questions in the style of forum moderator Chen Yiwei, checking standards text before it replies.
Audit Swimlane Flowchart Generator
nigo81/nigo-skills
Chinese-language tool that turns a business process description into a swimlane flowchart for internal control and audit work, exported as .drawio, PNG, SVG or VSDX.
Hidden Related-Party Audit Check
nigo81/nigo-skills
Cross-checks a client's own registry data against its customers' and suppliers' to surface undisclosed related-party relationships for an audit.
OpenCode Session Visualizer
nigo81/nigo-skills
Exports an OpenCode session from its local database into a single self-contained HTML page styled like the official share view, with folding and editing controls.
Related
Works with
Categories
Questions about CICPA Company Registration Query
What does CICPA Company Registration Query do?
Looks up Chinese company registration data in the CICPA industry knowledge base, from a quick search to a full 61-dimension export, subsidiary discovery and Excel output. cn, the research system of the Chinese Institute of Certified Public Accountants. It has four modes: a light single-company search, a detail view with shareholders, key personnel and equity structure, a basic batch query of 22 fields into one Excel file, and a full export of 61 dimensions packed as a ZIP of 53 Excel files.
When should I use CICPA Company Registration Query?
CICPA Company Registration Query fits situations like: checking a company's registered capital, legal representative or registered address; pulling a full company profile for due diligence or audit working papers; finding subsidiaries of a company by shareholding; batch-querying a list of companies into an Excel sheet.
How do I install CICPA Company Registration Query in Claude Code?
Run `npx skills add nigo81/nigo-skills --skill cicpa-company-query -a claude-code`. Or copy the skill folder (cicpa-company-query in nigo81/nigo-skills) into .claude/skills/cicpa-company-query in your project. Claude Code loads it when a task matches its description.
How do I install CICPA Company Registration Query in Codex?
Run `npx skills add nigo81/nigo-skills --skill cicpa-company-query -a codex`. Or copy the skill folder (cicpa-company-query in nigo81/nigo-skills) into .agents/skills/cicpa-company-query in your project. Codex loads it when a task matches its description.
Can I use CICPA Company Registration Query in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add nigo81/nigo-skills --skill cicpa-company-query -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cicpa-company-query, .gemini/skills/cicpa-company-query, .github/skills/cicpa-company-query and .opencode/skills/cicpa-company-query in your project.
What does CICPA Company Registration Query need to run?
Going by SKILL.md and its folder, CICPA Company Registration Query needs Python for the scripts in its folder and the command-line tools its instructions call (python3 and pip). Our summary lists: Python to run scripts/cicpa_query.py; A browser login to the CICPA knowledge base to obtain cookies.
Does CICPA Company Registration Query access the network?
SKILL.md contains no URLs. Its commands use pip, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is CICPA Company Registration Query safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does CICPA Company Registration Query use?
CICPA Company Registration Query is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does CICPA Company Registration Query use?
About 2.8k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to CICPA Company Registration Query?
Skills that share tags, products or a category with CICPA Company Registration Query: Three-Statement Model Builder (ginlix-ai/LangAlpha, 1.8k stars), Dcf Model (Wind-Alice/AliceMarket, 134 stars), Dcf Model (Luciole-Studio/Misaka-Agent, 171 stars) and Datapack Builder (w95/awesome-claude-corporate-skills, 244 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains CICPA Company Registration Query?
nigo81 (a GitHub user) maintains it in nigo81/nigo-skills, which has 133 GitHub stars. The repository holds 13 skills in this directory. The repository was last updated on August 23, 2026.
Source: nigo81/nigo-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.