Enables searching TikTok videos by hashtags and retrieving trending content, with video metadata extraction including URLs, descriptions, and engagement statistics (views, likes, shares, comments).
TikTok MCP Service
A Model Context Protocol service for TikTok video discovery and metadata extraction. This service provides a robust interface for searching TikTok videos by hashtags and retrieving trending content, with built-in anti-detection measures and error handling.
Features
- Search videos by hashtags
- Configurable video count per search (default: 30)
- Anti-bot detection measures
- Proxy support
- Automatic API session management
- Rate limiting and error handling
- Health status monitoring
Configuration
The service uses environment variables for configuration. Create a .env
file with:
Installation and Setup
Installing via Smithery
To install TikTok Video Search for Claude Desktop automatically via Smithery:
Installing Manually
Claude Desktop Integration
Once your service is running, you can integrate it with Claude Desktop. Since we're using Poetry for dependency management, make sure to run the MCP CLI commands through Poetry:
After installation, the service will be available in Claude Desktop and will run using Poetry for proper dependency management.
API Endpoints
Health Check
GET /health
- Check service health and API initialization status
Search Videos
POST /search
- Search for videos with hashtagsResponse includes video URLs, descriptions, and engagement statistics (views, likes, shares, comments).
Resource Management
POST /cleanup
- Clean up resources and API sessions
Error Handling
The service includes comprehensive error handling for:
- API initialization failures
- Bot detection issues
- Network errors
- Rate limiting
- Invalid search terms
Development
Built with:
- TikTokApi
- FastMCP
- Poetry for dependency management
- Playwright for browser automation
License
MIT# tiktok_mcp
TikTok API Limitations
Important Notice: TikTok has implemented strict anti-scraping measures that limit API access. As a result, this service provides the following functionality:
- Mock Data Mode: When TikTok blocks API access (which is currently the case), the service provides realistic-looking simulated results that are relevant to the search terms. This ensures that your Claude integration continues to function even when TikTok restricts access.
- API Access Attempts: The service still attempts to use the TikTok API first, but will quickly fall back to mock data if the API is unavailable or returns errors.
- Transparency: When mock data is provided, this is clearly indicated in the response via the
transformations
field, which includes a note explaining that simulated results are being shown.
This implementation ensures your service remains operational despite TikTok's anti-scraping measures.
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
Crowdlistening
Related MCP Servers
- Python
- -securityAlicense-qualityAn MCP server that tracks trending AI models, datasets, and spaces on Hugging Face.Last updated -PythonMIT License
- -securityAlicense-qualityA CLIP-Based Fashion Recommender system that allows users to upload clothing images and receive tags and recommendations based on visual analysis.Last updated -PythonApache 2.0
- TypeScript