{
  "markdown": "# DataSinking\n\n<!-- mcp-name: io.github.heubme2020/datasinking -->\n\n[![PyPI version](https://img.shields.io/pypi/v/datasinking.svg)](https://pypi.org/project/datasinking/)\n[![MCP server](https://img.shields.io/badge/MCP-server-blue)](https://github.com/heubme2020/datasinking#mcp-server)\n\n**Full-text financial reports across Asia, as clean Markdown.**\n\n[DataSinking](https://datasink.ing) serves **full-text financial reports** — annual, semi-annual\nand quarterly — from **China, Korea and Japan** as clean **Markdown**, ready for LLM reading\nand RAG. Query by FMP-style symbol (`600519.SS`, `005930.KS`, `7203.T`) or filter by exchange,\nreport period, or **section** — pull just the MD&A / risk section instead of the whole report.\nReports are sourced from official disclosure platforms and parsed into structured Markdown with\nYAML frontmatter, preserved headings, paragraphs and tables.\n\n---\n\n## MCP server\n\nShip DataSinking to any AI agent (Claude Desktop / Cursor / Codex / Windsurf) as an\n[MCP](https://modelcontextprotocol.io) server — 6 tools: list exchanges, list stocks,\nlist reports, fetch a report, list sections, fetch one section (token-friendly for RAG).\n\n```bash\npip install \"datasinking[mcp]\"\ndatasinking-mcp          # requires DATASINK_API_KEY (free at https://datasink.ing)\n```\n\nOr add to your client with `command: datasinking-mcp`. A remote streamable-HTTP endpoint\nis also live at `https://api.datasink.ing/mcp`. See [`mcp-server.md`](mcp-server.md).\n\n![DataSinking MCP in Claude](docs/images/mcp-demo.png)\n\n## What this repo is\n\nExamples, research and tutorials showing how to work with financial report data, including reproducing the presentation styles found in financial-report research papers.\n\n```\ndatasinking/\n├── examples/     # Example scripts: pull data from the API and analyze it\n├── research/     # Research notes / blog posts (reproducing paper-style presentation)\n├── datasinking/  # Python client + MCP server — pip install \"datasinking[mcp]\"\n├── mcp-server.md # How to configure the MCP server (for AI agents: Claude / Cursor / Codex / DeepSeek)\n├── llm-examples.md  # Ask an LLM — no code needed (8 end-to-end examples)\n├── api-examples.md  # 7 examples × 3 interfaces (curl / Python / LLM)\n└── README.md\n```\n\n## Quick start\n\n1. Get an API key at [datasink.ing](https://datasink.ing)\n2. One line (FMP-style `?apikey=`):\n\n```bash\ncurl \"https://api.datasink.ing/documents?symbol=600519.SS&with_content=1&apikey=YOUR_KEY\"\n```\n\nOr in Python:\n\n```bash\npip install datasinking\n```\n\n```python\nfrom datasinking import DataSinking\n\nds = DataSinking(\"YOUR_KEY\")\nfor r in ds.get_stock_reports(\"600519.SS\", limit=3):\n    print(r[\"report_period\"], r[\"title\"], len(r[\"content\"]), \"chars\")\n```\n\nAll five functions (curl / Python / LLM): [`api-examples.md`](api-examples.md).\n\n## Ask an LLM (no code)\n\nDon't want to write code? Point any LLM at [datasink.ing](https://datasink.ing),\ngive it your API key, and ask in plain language. See\n[`llm-examples.md`](llm-examples.md) for eight end-to-end examples — explore\ncoverage, list a company's reports, and extract a figure with correct units.\n\n## Examples (`examples/`)\n\n| File | What it does |\n|---|---|\n| `01_quickstart.py` | The 5 core functions: list exchanges / stocks / reports / fetch a report / fetch a stock's reports |\n| `02_download_company.py` | Download a company's full reports to local Markdown files |\n| `03_download_exchange.py` | Download an entire exchange's reports (all stocks) to local Markdown files |\n\nEvery example pulls from the live API and runs as-is.\n\n> `03_download_exchange.py` fetches every report on an exchange (e.g. all of Shenzhen — 150k+ documents). Free keys work too, but fall back to slow per-document fetching (1 req/s + shared daily quota); a **paid (yearly)** key is strongly recommended for full-exchange downloads.\n\n## Research (`research/`)\n\n`research/` hosts research notes and blog posts, each based on DataSinking data with the source cited. You can reproduce charts and presentations found in financial-report research papers, e.g.:\n\n- Long-term revenue / profit trends\n- Industry comparison and distribution\n- Time series of financial metrics\n\nStart from [`research/TEMPLATE.md`](research/TEMPLATE.md).\n\n## Data overview\n\n| | |\n|---|---|\n| Coverage | China (SSE / SZSE / BSE) · Korea (KOSPI / KOSDAQ / KONEX) · Japan (TSE) |\n| Document types | annual / semiannual / q1 / q3 / amendment |\n| Update frequency | Daily — Korea/Japan via official DART/EDINET APIs (new filings within ~24h of publication) |\n| Format | Full-text Markdown (with YAML frontmatter) |\n| API | REST — `GET /documents`, batch download, `with_content=1` for full text, `?section=` + `/sections` for chapter-level access |\n| Symbols | FMP style: `600519.SS` / `005930.KS` / `7203.T` |\n| Auth | `?apikey=` query parameter (FMP style) |\n\n## Data source\n\nReports are sourced from official regulatory disclosure platforms in each market and converted in-house to clean Markdown.\n\n## License\n\n[MIT](LICENSE)\n",
  "bytes": 4997,
  "sha": "9ab42a7f05ed3a95c9a1413a0ef8c04e05b1cf0ae6818be61d5c0f70d0a69543",
  "repo_slug": "heubme2020/datasinking",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_heubme2020_datasinking_5f8b336a/readme"
}