mcpField
T
TikTok Scraper (Apify) – videos, profiles, comments, transcripts

TikTok Scraper (Apify) – videos, profiles, comments, transcripts

Remote
apify.comSocial Media0Updated 1w ago

About TikTok Scraper (Apify) – videos, profiles, comments, transcripts

A hosted MCP server and agent skill that connects AI agents to the TikTok Scraper Apify Actor. It returns structured data on videos, profiles, comments, followers, sounds, and hashtags, with optional transcripts, downloads, and AI labels.

Key features

  1. 01Collect video metrics and creator profiles
  2. 02Retrieve comments and replies
  3. 03Read follower lists, sounds, and hashtag feeds
  4. 04Optionally provide transcripts, downloads, and AI labels

Use cases

  1. 01Review recent videos and engagement for creators
  2. 02Export comments and replies from a video
  3. 03Assess sponsorship and brand-safety signals

MCP support

The hosted MCP server exposes the Apify Actor to MCP clients. An agent skill describes input choices, defaults, paid options, and how to interpret result rows.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.apify.com/?tools=harshmaur/tiktok-scraper
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://mcp.apify.com/?tools=harshmaur/tiktok-scraperAuth required

Tools

Details

From the project README

TikTok Scraper — MCP server and agent skill

Let an AI agent read TikTok. This repo connects Claude, Cursor, Codex and other MCP clients or skill-aware agents to the TikTok Scraper Apify Actor.

Ask "what were the most liked iced matcha videos this week" or "vet @creator for brand safety", and the agent gets structured rows: videos with views, likes, comments, shares and saves; creator profiles with follower counts; comments with replies; follower and following lists; sounds and hashtag feeds. Optional extras are watermark-free MP4s, transcripts and AI labels for category, sponsored content, brand safety and sentiment.

No TikTok account, cookies or API key. You need only an Apify account.

Hosted MCP server

https://mcp.apify.com/?tools=harshmaur/tiktok-scraper

Sign in with OAuth in the client, or send Authorization: Bearer <APIFY_TOKEN> with a token from Apify Console → Integrations.

Claude Code:

claude mcp add --transport http tiktok https://mcp.apify.com/?tools=harshmaur/tiktok-scraper

Cursor, Windsurf, VS Code and others (mcp.json):

{
  "mcpServers": {
    "tiktok-scraper": {
      "url": "https://mcp.apify.com/?tools=harshmaur/tiktok-scraper"
    }
  }
}

Registry name: io.github.harshmaur/tiktok-scraper in the official MCP Registry.

Agent skill

skills/apify-tiktok-scraper/SKILL.md tells an agent which input list fits a request, that the per-target video count defaults to 1, which options cost extra, and how to read each row type.

Claude Code plugin (skill plus the hosted MCP tool):

/plugin install apify-tiktok-scraper --marketplace harshmaur/apify-tiktok-scraper

Any agent that reads skills:

npx skills add https://github.com/harshmaur/apify-tiktok-scraper

What you can ask

  • "The 30 newest videos from @khaby.lame with views and likes."
  • "Videos under #booktok since September with at least 50,000 likes, exact tag only."
  • "Export comments and replies from this video link."
  • "Check the last 20 videos of these five creators. Which look sponsored, and are they brand safe?"

Price

$0.35 per 1,000 videos, $0.20 per 1,000 comments, $0.25 per 1,000 followers, $4 per 1,000 profiles. Date and likes filters, exact hashtag match and TikTok's own captions are free. AI labels, AI transcripts and video downloads cost extra. Inputs that fail come back as free failed rows.

The input uses the same field names as clockworks/tiktok-scraper, so a saved input from there runs as it is.

Links

Disclosure: the Actor is paid per result and built by the author of this repo. Links carry no affiliate parameters.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.