Smithery MCP Registry Scraper

Scrape the Smithery Model Context Protocol (MCP) server registry. Returns server metadata, use counts, verification status, deployment info, and optionally the full tool/resource/prompt schemas for each server — covering 5,000+ entries and growing weekly.

---

Smithery MCP Registry Scraper Features

• Extracts the full Smithery catalog: 5,000+ MCP servers with name, description, use count, verification badge, and deployment URL
• Supports three modes: full catalog sweep, keyword search, or single-server detail lookup
• Retrieves per-server tool schemas, resource definitions, and prompt specs when includeTools is enabled
• Returns the connection type (HTTP, stdio) and config schema for each server
• No proxy required. Smithery's public API returns clean JSON without auth or browser rendering
• Handles page-based pagination automatically — point it at the catalog and walk away

---

What Can You Do With Smithery MCP Data?

• AI agent builders — Discover and index MCP servers that match your agent's domain, then wire in the right tools without manually browsing the directory
• IDE / editor integrations — Pull the full catalog to populate server-picker UIs in Cursor, Windsurf, Zed, or any MCP-capable client
• Competitive intelligence — Track which MCP servers are gaining use counts fastest, who's verified, and what tool schemas they expose
• Developer relations teams — Export the registry to find servers in your product category and reach out to their owners
• Security and compliance — Audit the full catalog for config schemas that accept sensitive parameters before onboarding servers to your platform

---

How Smithery MCP Registry Scraper Works

1. Choose a mode: full catalog, keyword search, or single-server detail
2. The scraper calls Smithery's public JSON API at registry.smithery.ai/servers with page-based pagination
3. For each server listing, it records the metadata from the catalog endpoint — name, use count, verified status, and deployment info
4. If includeTools is enabled, it fetches each server's detail endpoint separately to get tool schemas, resource definitions, and prompt specs

---

Input

{
  "mode": "catalog",
  "maxItems": 100,
  "includeTools": false,
  "verifiedOnly": false
}

Field	Type	Default	Description
mode	string	catalog	catalog (full sweep), search (keyword query), or server_detail (single server by qualified name)
query	string	""	Keyword to search — used in search mode
qualifiedName	string	""	Exact server identifier, e.g. exa or upstash/context7-mcp — used in server_detail mode
verifiedOnly	boolean	false	If true, limits results to Smithery-verified servers
includeTools	boolean	false	Fetch per-server detail (tools, resources, prompts) — adds one HTTP call per server
maxItems	integer	10	Max servers to return; 0 means no limit

Search mode example:

{
  "mode": "search",
  "query": "github",
  "maxItems": 50
}

Single-server detail example:

{
  "mode": "server_detail",
  "qualifiedName": "upstash/context7-mcp"
}

---

Smithery MCP Registry Scraper Output Fields

Catalog / search mode output:

{
  "qualified_name": "upstash/context7-mcp",
  "display_name": "Context7",
  "description": "Fetch up-to-date documentation and code examples directly into your prompts.",
  "icon_url": "https://api.smithery.ai/servers/upstash/context7-mcp/icon",
  "homepage_url": "https://github.com/upstash/context7",
  "verified": true,
  "use_count": 11216,
  "is_deployed": true,
  "by_smithery": false,
  "remote": true,
  "deployment_url": null,
  "tools_count": null,
  "tools": null,
  "connection_type": null,
  "server_url": "https://smithery.ai/server/upstash/context7-mcp",
  "created_at": "2025-04-12T07:00:21.424Z"
}

Field	Type	Description
qualified_name	string	Unique namespace/slug identifier on Smithery
display_name	string	Human-readable server name
description	string	Server description from the registry
icon_url	string	URL to the server's icon image
homepage_url	string	Server homepage or documentation URL
verified	boolean	Whether Smithery has verified this server
use_count	number	Cumulative installs reported by Smithery
is_deployed	boolean	Whether the server has a hosted remote deployment
by_smithery	boolean	Whether the server is published by Smithery itself
remote	boolean	Whether the server supports remote connections
deployment_url	string	Remote deployment URL when is_deployed is true
server_url	string	Direct URL to the Smithery server page
created_at	string	Creation timestamp (ISO 8601)

With includeTools: true, these fields are also populated:

Field	Type	Description
tools_count	number	Number of tools the server exposes
resources_count	number	Number of resources the server exposes
prompts_count	number	Number of prompts the server exposes
tools	array	Tool names and descriptions as "name|description" strings
resources	array	Resource names and URIs as "name|uri" strings
prompts	array	Prompt names and descriptions as "name|description" strings
connection_type	string	Primary connection type (http, stdio)
config_schema_json	string	JSON-encoded config schema for the connection

---

🔍 FAQ

How do I scrape the full Smithery MCP catalog?

Set mode to catalog and maxItems to 0. The scraper pages through the entire registry at pageSize=100 and collects every server. There are currently 5,000+ servers, which takes a few minutes at the default 200ms politeness delay.

How much does Smithery MCP Registry Scraper cost to run?

Smithery MCP Registry Scraper charges $0.10 to start plus $0.001 per record. Scraping 5,000 servers costs roughly $5.10. With includeTools: true, run time increases but the per-record charge stays the same.

What's the difference between catalog mode and server_detail mode?

Catalog mode pages through listing data — fast, good for bulk exports. The listing endpoint doesn't return tool or resource schemas. Use server_detail when you need a single server's full schema, or set includeTools: true in catalog or search mode to enrich every record with detail data.

Does Smithery MCP Registry Scraper need proxies?

No. Smithery's public registry API is open — no auth, no geo-restriction, and no bot detection beyond Cloudflare's CDN layer, which passes through without challenge on standard HTTP requests.

Can I filter to verified servers only?

Yes. Set verifiedOnly: true in any mode. Verified servers are a subset of the catalog that Smithery has reviewed and badged.

---

Need More Features?

Need additional fields, custom filters, or cross-registry comparison with Glama? File an issue or get in touch.

Why Use Smithery MCP Registry Scraper?

• First mover — The only Apify actor targeting the Smithery registry, which is the dominant MCP directory referenced directly in Anthropic, Cursor, and Windsurf documentation
• Clean output — Returns structured JSON with consistent field names, including optional per-server tool and resource schemas not available from any other programmatic source
• No overhead — Pure JSON API scraping, no browser required, no proxies needed, runs in 256MB memory