Skip to main content
Glama
README.md
# @pipeworx/china-tax-policy

China's tax policy documents from the State Taxation Administration's policy library: tax laws, departmental rules, STA normative documents and MOF/STA joint announcements, in the original Chinese text with document number, issuing bodies and dates.

Part of [Pipeworx](https://pipeworx.io) — an MCP gateway connecting AI agents to 1880+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

## Tools

- `cn_tax_policy_search(query?, category?, issuer?, year?, number?, page?, limit?)` — search by Chinese keywords, or look a document up by its number. A query such as `2023年第39号` (or `year` + `number`, optionally `issuer` such as `财政部税务总局公告`) resolves directly to that document. Returns title, document number (文号), issuing bodies, document date (成文日期), category, tax types, the publisher's validity flag (全文废止 / 已修改, null when unmarked) and an `id`.
- `cn_tax_policy_text(id)` — the full original text of one document from the search, with its document number, document date and the effective date stated in the text when it gives one (for example "自2023年8月28日起").

## Auth

Keyless.

## Data sources

- State Taxation Administration (国家税务总局), tax policy and regulation library (政策法规库) — laws, administrative regulations, departmental rules, normative documents and MOF/STA notices on tax. Each call goes to the publisher live; nothing is stored here.

Notes for the next person:

- `document_date` is the date the document was written (成文日期). The effective date is only in the text, so it comes from `effective_date_stated_in_text`, which is null when the text names none (many documents take effect on issue or on a date that needs reading in context).
- `validity_flag` is the publisher's own marking from its index. Null means the publisher marks nothing, not that the document is in force.
- The source returns ten results per page whatever page size is asked for; `limit` trims within the page and `page` moves through the rest.
- Search keywords are segmented by the publisher, so broad queries return loosely related documents. Use the document number when you have one.
- Policy interpretations (政策解读) and Q&A in the same library are not covered; this pack serves the documents themselves.

## Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

```json
{
  "mcpServers": {
    "china-tax-policy": {
      "url": "https://gateway.pipeworx.io/china-tax-policy/mcp"
    }
  }
}
```

### What this endpoint actually serves

`tools/list` at `https://gateway.pipeworx.io/china-tax-policy/mcp` returns the tools in the table
above **plus the shared Pipeworx meta-tools** — `ask_pipeworx`,
`discover_tools`, `search_within`, `remember`/`recall` and the rest of the
gateway-wide set. So the tool count you see is larger than this table: a
single-pack endpoint currently lists roughly 30 shared tools alongside the
pack's own. The connection's `initialize` response states its exact scope, and
is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a
scoped connection answer a question this pack does not cover — via
`ask_pipeworx`, which routes across the whole catalog — without you adding a
second MCP server. There is currently no way to mount a pack endpoint without
them; if the extra schemas cost you more context than the routing is worth,
connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed
directly, instead of just this one's:

```json
{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}
```

Both URLs reach the same gateway and the same 1880+ data sources. The
only difference is which pack's tools are listed **directly**; `ask_pipeworx`
reaches all of them from either one.

## No MCP client? Call it over HTTP

```bash
curl -X POST https://gateway.pipeworx.io/v1/tools/cn_tax_policy_search \
  -H 'Content-Type: application/json' \
  -d '{"query":"2023年第39号","issuer":"财政部税务总局公告"}'
```

No account needed for the first calls. Inspect any tool: `GET https://gateway.pipeworx.io/v1/tools/cn_tax_policy_search`. Find one: `POST https://gateway.pipeworx.io/v1/tools/search_packs` with `{"query":"..."}`.

## Run it from npm

`@pipeworx/mcp-china-tax-policy` is a local stdio bridge to the Pipeworx gateway: your MCP client
starts it with `npx`, and it forwards every call to the same
`https://gateway.pipeworx.io/china-tax-policy/mcp` endpoint shown above. It needs network access;
there is no offline mode.

```json
{
  "mcpServers": {
    "china-tax-policy": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-china-tax-policy"]
    }
  }
}
```

Without a key it calls anonymously, capped at 50 calls a day. A free account
raises that to 200 a day: get a key at https://pipeworx.io/signup and add
`"env": { "PIPEWORX_API_KEY": "<your key>" }` to the server entry.
The key is sent as a Bearer token, the same way a hosted connection sends it,
so calls are metered and billed exactly as they are there.

It lists **only** this pack's tools by default. Set `PIPEWORX_META_TOOLS=1` to
also list `ask_pipeworx` and the other gateway-wide tools.

Behind an HTTPS proxy: Node's built-in `fetch` does not read `HTTPS_PROXY` by
default. Newer Node releases honour it when `NODE_USE_ENV_PROXY=1` is set; check
your Node version's documentation.

## Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English —
this works on the pack endpoint above as well as on the full gateway:

```
ask_pipeworx({ question: "your question about China Tax Policy data" })
```

The gateway picks the right tool and fills the arguments automatically.

## More

- [Docs and guides](https://pipeworx.io/docs)
- [pipeworx.io](https://pipeworx.io)

## License

MIT