MCP Server for Taiwan Company Registry — AI-callable tools for querying GCIS open data via Model Context Protocol
Project description
MCP Taiwan Company Registry
An MCP server for Taiwan company registry open data, exposing AI-callable tools over Model Context Protocol (MCP). Powered by the Ministry of Economic Affairs GCIS open data endpoints.
Features
- stdio JSON-RPC 2.0 — Standard MCP transport protocol
@mcp.tool()decorator — Pydantic-typed tool registration- REST connector — Retry support with exponential backoff
- No-auth public API access — Uses the GCIS open data endpoint directly
- E2E testing — Live API test runner
- 9 registered tools — Company search, directors, business items, branch offices, and more
Prerequisites
- Python
3.14.x uv- No API key required (public open data)
Installation
From PyPI
pip install mcp-tw-company
With uvx (no install needed)
uvx mcp-tw-company
From source
git clone https://github.com/asgard-ai-platform/mcp-tw-company.git
cd mcp-tw-company
uv sync
Configuration
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"tw-company": {
"command": "uvx",
"args": ["mcp-tw-company"]
}
}
}
Claude Code
Add to .mcp.json in your project:
{
"mcpServers": {
"tw-company": {
"command": "uvx",
"args": ["mcp-tw-company"]
}
}
}
Cursor
Add to Cursor MCP settings:
{
"mcpServers": {
"tw-company": {
"command": "uvx",
"args": ["mcp-tw-company"]
}
}
}
Quick Start
# Setup
uv sync
# Test connection
uv run python scripts/auth/test_connection.py
# Run server
uv run python mcp_server.py
Usage Examples
"Search for companies by name"
You: 幫我查「宏碁」相關的公司
AI calls:
get_company_registrations_by_name_keyword(
company_name_keyword = "宏碁",
company_status = "01",
)
Result: SUCCESS — Returns matching company registrations including name, unified business number, capital, representative, and address.
"Look up a specific company by number"
You: 統一編號 89845559 是哪家公司?
AI calls:
get_company_registration_master_by_number(
business_accounting_no = "89845559",
)
Result: SUCCESS — Returns master registration profile: 森鉅科技材料股份有限公司, capital NT$3B, located in Tainan.
"Find directors and supervisors"
You: 查一下這家公司的董監事名單
AI calls:
get_company_directors_by_number(
business_accounting_no = "89845559",
)
Result: SUCCESS — Returns 9 directors/supervisors with positions, names, corporate representatives, and shareholding.
"Check what business items a company is registered for"
You: 這家公司的營業項目有哪些?
AI calls:
get_company_registration_with_business_items_by_number(
business_accounting_no = "89845559",
)
Result: SUCCESS — Returns company profile with 13 registered business items including manufacturing, trading, and construction.
"Determine entity type by number"
You: 統編 00208407 是公司還是商號?
AI calls:
get_registry_entity_type_by_number(
no = "00208407",
)
Result: SUCCESS — Returns entity classification (company, branch office, or business).
Tools Reference
| Tool | Description | Key Parameters |
|---|---|---|
get_company_registrations_by_name_keyword |
Search companies by name keyword | company_name_keyword, company_status |
get_company_registration_master_by_number |
Company master profile | business_accounting_no (8 digits) |
get_company_registration_with_business_items_by_number |
Company profile with business items | business_accounting_no (8 digits) |
get_company_directors_by_number |
Directors and supervisors | business_accounting_no (8 digits) |
get_company_name_by_number |
Company name lookup | business_accounting_no (8 digits) |
get_registry_entity_type_by_number |
Entity type classification | no (8 digits) |
get_branch_offices_by_company_number |
Branch offices by company | business_accounting_no (8 digits) |
get_business_registration_profile_by_agency |
Business profile by agency | president_no, agency |
get_business_registration_items_by_number |
Business registration items | president_no (8 digits) |
All tools support skip (default 0) and top (default 50) pagination parameters.
TODO
- Integrate
company registration basic data (application 2)fromF05D1060-7D57-4763-BDCE-0DAF5975AFE0 - Integrate
business name by numberfrom855A3C87-003A-4930-AA4B-2F4130D713DC - Investigate the upstream access restriction for
business registrations by name keyword(A1B4CBFF-2D3A-409B-8A78-2AD94F63AE4A): the test endpoint currently returnsunauthorized integration IPin this environment - Investigate the upstream access restriction for
branch offices by branch number(23632BB3-5DB7-4423-9643-1D4AC140D479): the test endpoint currently returnsunauthorized integration IPin this environment
Project Structure
mcp-tw-company/
├── app.py # FastMCP singleton
├── mcp_server.py # Entry point (stdio transport)
├── config/settings.py # API endpoints, URL builder, request headers
├── connectors/
│ └── rest_client.py # HTTP REST with retry + pagination
├── auth/
│ └── none.py # No auth (public API)
├── tools/
│ └── company_registry_tools.py
├── tests/test_all_tools.py # E2E test runner
└── scripts/auth/test_connection.py
Testing
uv run python scripts/auth/test_connection.py # Validate API connectivity
uv run python tests/test_all_tools.py # Run all tool E2E tests
Data Source
This project uses Ministry of Economic Affairs GCIS open data endpoints for company and business registration data.
Contributing
See CONTRIBUTING.md for development setup and guidelines.
License
MIT License — see LICENSE for details.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file mcp_tw_company-0.1.0.tar.gz.
File metadata
- Download URL: mcp_tw_company-0.1.0.tar.gz
- Upload date:
- Size: 72.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
a68e846d91539f7fc7aaf12606141b32c1d52c38b97a1462073bcb67b60e24af
|
|
| MD5 |
f8a51c3776a121c2028fe456f0591183
|
|
| BLAKE2b-256 |
163fc79250a7ea7fcd2f3754ba45a32ce2f62d9d06b96ac8a72c78c56cf23826
|
Provenance
The following attestation bundles were made for mcp_tw_company-0.1.0.tar.gz:
Publisher:
publish.yml on asgard-ai-platform/mcp-tw-company
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
mcp_tw_company-0.1.0.tar.gz -
Subject digest:
a68e846d91539f7fc7aaf12606141b32c1d52c38b97a1462073bcb67b60e24af - Sigstore transparency entry: 1295734073
- Sigstore integration time:
-
Permalink:
asgard-ai-platform/mcp-tw-company@1c4b1716d3fa3bc528981a652e18adec0f889c56 -
Branch / Tag:
refs/tags/0.1.0 - Owner: https://github.com/asgard-ai-platform
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@1c4b1716d3fa3bc528981a652e18adec0f889c56 -
Trigger Event:
release
-
Statement type:
File details
Details for the file mcp_tw_company-0.1.0-py3-none-any.whl.
File metadata
- Download URL: mcp_tw_company-0.1.0-py3-none-any.whl
- Upload date:
- Size: 12.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
5870584763c3cf7a94087f9c80af509f6023072682ef3bb5a7bd02535d5bc78c
|
|
| MD5 |
990cfca52a0d5a0cec92e67f5929e5b3
|
|
| BLAKE2b-256 |
b077d15f1d872c3189d313d84c3c8dc5928acf14fed3279a75783b32e8839725
|
Provenance
The following attestation bundles were made for mcp_tw_company-0.1.0-py3-none-any.whl:
Publisher:
publish.yml on asgard-ai-platform/mcp-tw-company
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
mcp_tw_company-0.1.0-py3-none-any.whl -
Subject digest:
5870584763c3cf7a94087f9c80af509f6023072682ef3bb5a7bd02535d5bc78c - Sigstore transparency entry: 1295734157
- Sigstore integration time:
-
Permalink:
asgard-ai-platform/mcp-tw-company@1c4b1716d3fa3bc528981a652e18adec0f889c56 -
Branch / Tag:
refs/tags/0.1.0 - Owner: https://github.com/asgard-ai-platform
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@1c4b1716d3fa3bc528981a652e18adec0f889c56 -
Trigger Event:
release
-
Statement type: