# Douyin_TikTok_Download_API: Self-Hosted No-Watermark Scraper & MCP Server

> Douyin_TikTok_Download_API offers a self-hosted Python MCP Server for scraping Douyin and TikTok, downloading no-watermark videos, and archiving content. It solves the problem of platform data collection and media preservation with an identity pool, PostgreSQL archive, and one-command Docker deployment. Developers needing robust, self-managed access to TikTok/Douyin data via MCP will find it useful.

**Published:** 2026-09-17T12:00:11.823+00:00

**Keywords:** douyin-tiktok-download-api,mcp-server,tiktok-scraper,douyin-api

# Douyin_TikTok_Download_API: Self-Hosted No-Watermark Scraper & MCP Server

Douyin_TikTok_Download_API provides a self-hosted Python MCP Server, alongside a REST API, CLI, and web console, for comprehensive data collection and no-watermark video downloads from Douyin and TikTok. It allows developers to parse links, share text, or post IDs, archive collected content to PostgreSQL, and download media in bulk, skipping duplicates. The project, with over 20,000 GitHub stars, emphasizes self-reliance with an identity pool that maintains itself and a simple `docker compose up` deployment.

## Integrating via MCP

The Douyin_TikTok_Download_API exposes its core service layer directly through an MCP endpoint at `/mcp`. This means that the same capabilities available via the REST API and CLI—parsing, archiving, downloading, and watching authors or posts—are accessible through the Model Context Protocol. For client setup and guidance on connecting, the server provides a dedicated page at `/mcp-guide` within its web console. This direct integration ensures that developers can leverage the protocol for structured interaction with the scraper's functionalities.

## Access Points and Capabilities

Beyond the MCP endpoint, the Douyin_TikTok_Download_API offers several ways to interact with its features:

*   **Web console**: Accessible at the root path (`/`), it provides a UI for managing the identity pool, scheduler, library, downloads, logs, and diagnostics.
*   **REST API**: Available under `/api/v1/...`, this offers 93 operations. API documentation is provided via Swagger UI at `/docs` (English and Chinese) and bare references at `/swagger` and `/redoc` (no login required).
*   **CLI**: The `dtk` command-line interface provides access to the same functionalities.

Core capabilities include:

*   **Parsing**: Process various input formats like links, share text, short links, or bare post IDs.
*   **Archiving**: Store all parsed content, ensuring data persistence even if the original post is deleted upstream.
*   **Downloading**: Save media to local disk, with bulk download options by author, duplicate skipping, and cleanup. The system retrieves the clean, watermark-free stream directly published by the platform.
*   **Watching**: Monitor specific authors or posts and re-collect their content on a timer.
*   **iOS Shortcut support**: An endpoint at `/api/v1/ios/shortcut` facilitates integration with iOS Shortcuts.

## Self-Hosted Data Management

The project is designed for complete self-hosting, eliminating external dependencies like sign-ups or quotas. It maintains its own identity pool for platform interaction and stores all collected data, including videos, image albums, posts, authors, comments, and search results, in a PostgreSQL database. This architecture provides developers with full control over their collected data and operational environment.

## References

*   [Douyin_TikTok_Download_API on GitHub](https://github.com/Evil0ctal/Douyin_TikTok_Download_API)
*   [Model Context Protocol Documentation](https://modelcontextprotocol.io/introduction)
*   [Douyin_TikTok_Download_API on model-context-protocol.com](https://model-context-protocol.com/servers/)

## Related Repository

- [douyin-tiktok-download-api](https://model-context-protocol.com/servers/douyin-tiktok-download-api)

**Source:** https://model-context-protocol.com/blog/douyintiktokdownloadapi-self-hosted-no-watermark-scraper-mcp-mcp-server-guide