MCP Servers Scraper — Official MCP Registry Export

Get the full list of MCP servers from the official Model Context Protocol registry (registry.modelcontextprotocol.io) as clean rows: server name, title, description, version, remote endpoints (streamable HTTP / SSE URLs), packages (npm, PyPI, Docker), status and publish/update dates. The registry lists close to 10,000 servers; this Actor exports all of them, or just the ones you filter for.

What does this MCP registry scraper do?

It pages through the registry's public GET /v0/servers API with cursors and writes one dataset row per server (latest version only by default). Filter by status, a name substring or a search term. No authentication, scraping or browser is needed, and the default input finishes quickly.

Why use it?

• Official registry API — no scraping, no auth
• Cursor-based pagination for full dumps
• Filter by status / name / latest-only
• Fail-loudly empty-result guard

Data source

Official MCP Registry API GET /v0/servers at https://registry.modelcontextprotocol.io (public).

Input

• status: active | deprecated | deleted | empty
• nameContains: Substring filter
• latestOnly: Only latest versions
• maxItems: Cap
• registryBase: Override base URL

Example input:

{
  "status": "active",
  "latestOnly": true,
  "maxItems": 200
}

Output

Each dataset item looks like this:

{
  "name": "io.github.modelcontextprotocol/server-github",
  "title": "GitHub",
  "version": "1.0.0",
  "status": "active",
  "remotes": []
}

Export as JSON, CSV, Excel or HTML, or pull via the Apify API, webhooks, Make, Zapier, n8n or the Apify MCP server.

How much does it cost?

Pay per event: $0.001 per run + $0.001 per result (≈ $1.00 per 1,000 results). You are never charged for results on a failed run: if no data is found, the run fails loudly with an explanation instead of returning an empty "success".

Use cases

• MCP directories / marketplaces syncing catalog
• Agent builders discovering tools
• Researchers measuring MCP ecosystem growth
• Change-detection pipelines (diff two runs)

Scheduling & monitoring

Schedule it daily or weekly in Apify Console and connect a webhook/Slack/email integration to get new items automatically.

FAQ

Is this legal? It uses the official public MCP Registry HTTP API. Server metadata is published by authors to the registry.. No personal data is collected.

Something broke? Open an issue on the Issues tab; the actor is tested daily.

Tips

Tip: spot new MCP servers every day. Schedule a daily run and compare datasets (or filter on publishedAt) to see which MCP servers were added or updated since yesterday. Directories and agent marketplaces use this to stay in sync with the official registry.

More AI-ecosystem Actors from Agent Data Co

Built and maintained by Agent Data Co. Our Actors collect public AI-ecosystem data and work well together on a daily schedule:

• GitHub Trending AI Repos Scraper: AI agent, MCP and LLM repos with stars and topics
• Show HN Scraper (Hacker News AI launches): new AI launches on Hacker News with points and comments
• Hugging Face Scraper: trending models, datasets and Spaces
• LLM Pricing Scraper (OpenRouter): 400+ LLM prices with price-change detection
• SaaS Pricing Page Monitor: plan and price changes on any pricing page
• App Store Reviews Scraper: iOS reviews for any app, by country