# Seym0n/tiktok-mcp [Health: Active]

**Category:** 🛠️ Other Tools and Integrations  
**Repository:** https://github.com/Seym0n/tiktok-mcp  
**GitHub Stars:** 196  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/seym0n-tiktok-mcp

## Description
Model Context Protocol (MCP) with TikTok integration

## Tools
Capabilities this server exposes over MCP:

- **tiktok_url**
- **language_code**
- **query**
- **cursor**
- **search_uid**

## Claude Desktop Quick Installation
Heuristic fallback — verify the package name and runner against the repository README before running it. Uses `npx` (confidence: low):

```json
"mcpServers": {
  "tiktok-mcp": {
    "command": "npx",
    "args": ["-y","seym0n-tiktok-mcp"],
    "env": {
      "TIKNEURON_MCP_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `TIKNEURON_MCP_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What Seym0n/tiktok-mcp MCP server does

Seym0n/tiktok-mcp MCP server gives an MCP client access to selected TikTok data through TikNeuron. It supports three main operations: retrieving a video's subtitle, retrieving detailed information about a post, and searching for videos.

Subtitle retrieval accepts a TikTok URL and can optionally request a language code. Without a language code, the returned content comes from automatic speech recognition. Post details can be requested with a full TikTok URL or a video ID. The response may include the video's description, ID, creator username, hashtags, likes, shares, comments, views, bookmarks, creation date, duration, and available subtitle languages and sources.

Search accepts a text query such as a topic or content description. Results include video details and pagination metadata, allowing a client to continue a search with a cursor and search session identifier.

## How it works

Seym0n/tiktok-mcp MCP server runs as a Node.js MCP process and uses the TikNeuron service for TikTok access. The repository requires Node.js 18 or newer. A local setup clones the repository, installs its npm dependencies, and builds the project. The build produces `build/index.js`, which is then started by the MCP client with Node.

The README shows Claude Desktop configuration using the `node` command and the built JavaScript file. The configuration passes the TikNeuron credential through the `TIKNEURON_MCP_API_KEY` environment variable. The project also provides an MCPB release bundle that Claude Desktop can install through its graphical installation flow.

## Setup and configuration

The supported local build sequence is:

1. Clone the GitHub repository.
2. Run `npm install`.
3. Run `npm run build`.
4. Configure the client to launch the resulting `build/index.js` with Node.js.
5. Set `TIKNEURON_MCP_API_KEY` to the API key from a TikNeuron account.

For Claude Desktop, the README also documents downloading the latest `tiktok-mcp.mcpb` file from the project's releases and opening it in Claude Desktop. The installation dialog requests the TikNeuron API key. Git is listed as a requirement for the repository-based setup.

## Tools and capabilities

Seym0n/tiktok-mcp MCP server exposes these documented operations:

- `tiktok_get_subtitle`: obtains subtitle or recognized speech content from a TikTok video, with optional language selection.
- `tiktok_get_post_details`: returns post metadata, creator information, engagement counts, timing data, and subtitle availability.
- `tiktok_search`: searches for videos and returns matching video data plus pagination information.

The input fields described by the project include `tiktok_url`, `language_code`, `query`, `cursor`, and `search_uid`. URL inputs may use standard TikTok links or shortened `vm.tiktok.com` links; post details also accept a video ID.

## Limitations and notes

The material describes access through TikNeuron and requires a TikNeuron account and MCP API key. It does not document unauthenticated operation, alternate providers, rate limits, or a hosted endpoint. The README explicitly demonstrates Claude Desktop, while the project description also refers to other applications that support MCP. The repository instructions do not provide a package-manager command for installing and launching the project directly; they describe building it locally or using the MCPB release bundle.

_Full upstream README: https://allmcps.com/mcp/seym0n-tiktok-mcp/readme_