API Reference

curl

Every request carries your apikey.

1. List exchanges

curl "https://api.datasink.ing/exchanges?apikey=YOUR_API_KEY"

2. List stocks in an exchange

# China — Shanghai
curl "https://api.datasink.ing/stocks?exchange=sse&apikey=YOUR_API_KEY"

# Japan — Tokyo
curl "https://api.datasink.ing/stocks?exchange=jpx&apikey=YOUR_API_KEY"

# Korea — KOSDAQ
curl "https://api.datasink.ing/stocks?exchange=koe&apikey=YOUR_API_KEY"

# Taiwan — Taipei
curl "https://api.datasink.ing/stocks?exchange=twse&apikey=YOUR_API_KEY"

3. List a stock's reports (metadata)

# China — Moutai (600519.SS)
curl "https://api.datasink.ing/documents?symbol=600519.SS&doc_type=annual&apikey=YOUR_API_KEY"

# Japan — Toyota (7203.T)
curl "https://api.datasink.ing/documents?symbol=7203.T&doc_type=annual&apikey=YOUR_API_KEY"

# Korea — Samsung Electronics (005930.KS)
curl "https://api.datasink.ing/documents?symbol=005930.KS&doc_type=annual&apikey=YOUR_API_KEY"

# Taiwan — TSMC (2330.TW)
curl "https://api.datasink.ing/documents?symbol=2330.TW&doc_type=annual&apikey=YOUR_API_KEY"

4. Fetch one report (full text)

curl "https://api.datasink.ing/documents/42?apikey=YOUR_API_KEY"

5. Fetch a stock's reports (full text)

By reporting period (report_period, not disclosure date), or the latest N by order=desc&size=N:

# latest 7 (full text)
curl "https://api.datasink.ing/documents?symbol=600519.SS&order=desc&size=7&with_content=1&apikey=YOUR_API_KEY"

# by reporting-period range (full text)
curl "https://api.datasink.ing/documents?symbol=600519.SS&report_period_from=2023-01-01&report_period_to=2023-12-31&with_content=1&apikey=YOUR_API_KEY"

6. Fetch a single section (for RAG)

Reports are chapter-structured. List the sections, then pull just one — no need to download the whole report. E.g. pull the 管理层讨论与分析 (MD&A) chapter, or 财务报告 to get the notes (附注) and find a specific line item like 在建工程.

# list a report's sections
curl "https://api.datasink.ing/documents/3/sections?apikey=YOUR_API_KEY"

# pull one section only (fuzzy title match; non-ASCII titles must be URL-encoded)
curl "https://api.datasink.ing/documents/3?section=%E7%AE%A1%E7%90%86%E5%B1%82%E8%AE%A8%E8%AE%BA%E4%B8%8E%E5%88%86%E6%9E%90&apikey=YOUR_API_KEY"

# the notes (附注) live inside the financial-report chapter
curl "https://api.datasink.ing/documents/3?section=%E8%B4%A2%E5%8A%A1%E6%8A%A5%E5%91%8A&apikey=YOUR_API_KEY"

Python

pip install datasinking, then five functions:

from datasinking import DataSinking

ds = DataSinking("YOUR_API_KEY")

ds.list_exchanges()                          # -> ['bj', 'jpx', 'knx', 'koe', 'ksc', 'sse', 'szse', 'tpex', 'twse']
ds.list_stocks("sse")                        # -> list[dict], metadata
ds.list_reports("600519.SS", doc_type="annual")  # -> list[dict], metadata (no content)
ds.get_report(42)                            # -> dict, single report (full text)

# -> list[dict], full text. limit: latest N, -1 = all.
ds.get_stock_reports(
    "600519.SS",
    period_from="2023-01-01",   # by reporting period, not disclosure date
    period_to="2023-12-31",
    limit=7,                    # latest 7; -1 = all
    doc_type=None,
)

View on PyPI · Full source and examples on GitHub.

Using it from an LLM

Point any LLM at this site, give it your key, and ask in plain language — the model reads the API, calls it, and hands you the answer. Each of the five operations maps to a natural-language ask:

You have access to DataSinking — an API for full-text financial reports (annual,
semi-annual, quarterly) from the US, the UK, China, Japan, Korea and Taiwan, as clean Markdown.
Docs: https://datasink.ing/docs · Spec: https://api.datasink.ing/openapi.json
My API key: YOUR_API_KEY  (pass it as ?apikey=)

Workflow: list reports to get an id, then fetch by that id.
- List exchanges:  GET /exchanges
- List stocks:     GET /stocks?exchange=sse
- List reports:    GET /documents?symbol=600519.SS&doc_type=annual
                    → each item has an "id"
- Get one report:  GET /documents/{id}  (full Markdown)
- Get one section: GET /documents/{id}?section=%E7%AE%A1%E7%90%86%E5%B1%82%E8%AE%A8%E8%AE%BA%E4%B8%8E%E5%88%86%E6%9E%90  (= 管理层讨论与分析, URL-encoded — non-ASCII headings MUST be encoded; token-efficient for RAG)

Symbols are FMP-style: AAPL (Apple), 600519.SS (Moutai), 7203.T (Toyota), 005930.KS (Samsung).
Example: "Moutai's 2023 revenue" → list reports (doc_type=annual) → find 2023 → get that
report's section 主要财务指标 → read revenue with its unit (元/RMB).

The returned content opens with YAML frontmatter (identity + period) and marks unit rows like > 单位：元, so the model reads each figure at the correct scale. Full prompts: llm-examples.md. A machine-readable spec is at https://api.datasink.ing/openapi.json.

MCP server (for AI agents)

Connect Claude, Cursor, OpenAI Codex, DeepSeek or any MCP-compatible client directly via Model Context Protocol — ask in plain language, no code. Per-client setup for Claude Code, Claude Desktop, Cursor, Codex, WorkBuddy, Doubao Work, Devin Desktop and DeepSeek Harness →

Hosted — nothing to install

{
  "mcpServers": {
    "datasinking": {
      "type": "http",
      "url": "https://api.datasink.ing/mcp?apikey=YOUR_API_KEY"
    }
  }
}

Your key travels inside the URL, so keep this config private. Clients that support custom headers can send Authorization: Bearer YOUR_API_KEY instead.

Local — run it yourself

{
  "mcpServers": {
    "datasinking": {
      "command": "uvx",
      "args": ["--from", "datasinking[mcp]", "datasinking-mcp"],
      "env": { "DATASINK_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Runs via uvx — nothing to install either. Prefer a permanent install? Run pip install "datasinking[mcp]" and set "command" to "datasinking-mcp".

Prefer Node over Python? The same server ships on npm — no Python, no uv:

{
  "mcpServers": {
    "datasinking": {
      "command": "npx",
      "args": ["-y", "datasinking-mcp"],
      "env": { "DATASINK_API_KEY": "YOUR_API_KEY" }
    }
  }
}

Claude Code: claude mcp add --transport http datasinking https://api.datasink.ing/mcp --header "Authorization: Bearer YOUR_API_KEY"

Tools include get_section — pull one chapter of a report (token-efficient, ideal for RAG). Full setup guide: mcp-server.md.

Fields returned

Every document object carries these fields (the list endpoint wraps them in items with total / page / size):

content opens with a YAML frontmatter block (stock_code, stock_name, report_period, announcement_date, doc_type, title), so an LLM can read the document identity and reporting period without guessing from the text.