mcpservers APImcpservers.org ↗
Access MCP server listings, client apps, remote servers, and Claude skills from mcpservers.org. Search, filter by category, and retrieve full server details.
What is the mcpservers API?
This API exposes 9 endpoints covering the full mcpservers.org catalog — MCP servers, clients, remote servers, and Claude skills. Use search_servers to find servers by keyword, get_servers_by_category to filter by slugs like web-scraping or database, and get_server_detail to pull per-server metadata including GitHub URL, star count, and rendered README HTML for a specific repository slug.
curl -X GET 'https://api.parse.bot/scraper/54bddc77-e1be-4c3a-9f8d-27fd16c8d932/list_servers?page=1&sort=newest' \ -H 'X-API-Key: $PARSE_API_KEY'
Typed, relational, agent-ready
A generated client with real types, enums, and the links between objects — the structure a flat JSON response can't carry. Autocompletes in your editor and reads cleanly to coding agents.
- Fully typed · autocompletes
- Objects link to objects
- Typed errors & pagination
Typed Python client. Set up the SDK in your uv project, then pull this API’s typed client:
uv add parse-sdk uv run parse init uv run parse add --marketplace mcpservers-org-api
uv run parse add --marketplace pulls a pinned snapshot of this canonical API — it won’t change underneath you. To customize it, subscribe and swap to your own copy.
"""Walkthrough: Awesome MCP Servers API — discover servers, drill into details, explore the ecosystem."""
from parse_apis.awesome_mcp_servers_api import McpServers, Sort, ServerNotFound
client = McpServers()
# List newest servers with sorting
for server in client.servers.list(sort=Sort.NEWEST, limit=5):
print(server.name, server.badges)
# Search for database-related servers, drill into the first result's details
server = client.servers.search(query="database", limit=1).first()
if server:
detail = server.details()
print(detail.name, detail.description, detail.github_url)
# Browse by category — web-scraping servers
for s in client.servers.list_by_category(category="web-scraping", limit=3):
print(s.name, s.description[:60])
# Typed error handling for a non-existent server
try:
detail = client.servers.search(query="nonexistent-xyz-12345", limit=1).first()
except ServerNotFound as exc:
print(f"Server not found: {exc.slug}")
# Explore remote servers (OAuth/cloud-hosted)
for remote in client.remoteservers.list(limit=3):
print(remote.name, remote.slug)
# List Claude agent skills
for skill in client.skills.list(limit=3):
print(skill.name, skill.author)
print("exercised: servers.list / servers.search / server.details / servers.list_by_category / remoteservers.list / skills.list")
List all MCP servers with pagination and optional sorting. Returns 30 results per page. Servers include sponsors (no slug) and community/official entries (with slug). Paginate by incrementing page; total_servers indicates the full catalog size.
| Param | Type | Description |
|---|---|---|
| page | integer | Page number (30 results per page). |
| sort | string | Sort order for listings. |
{
"type": "object",
"fields": {
"page": "integer, current page number",
"sort": "string or null, active sort order",
"servers": "array of server objects with name, description, url, slug, and badges",
"total_servers": "integer or null, total number of servers available"
},
"sample": {
"data": {
"page": 1,
"sort": null,
"servers": [
{
"url": "https://get.brightdata.com/mcpservers",
"name": "Bright Data",
"slug": null,
"badges": [
"sponsor"
],
"description": "Discover, extract, and interact with the web."
},
{
"url": "https://mcp.alphavantage.co",
"name": "Alpha Vantage MCP Server",
"slug": null,
"badges": [
"sponsor"
],
"description": "Access financial market data."
}
],
"total_servers": 8959
},
"status": "success"
}
}About the mcpservers API
Server Discovery and Browsing
list_servers returns paginated server listings (30 per page) with optional sort=newest ordering. Each server object includes name, description, url, slug, and badges. get_featured_servers returns the homepage-featured subset without pagination parameters, and get_official_servers isolates the officially-curated set with page-based pagination.
Search and Category Filtering
search_servers accepts a required query string and an optional page integer, returning matching servers with the same object shape. get_servers_by_category takes a category slug — such as web-scraping, database, or search — and returns the page, category, and matching servers array. Both endpoints are useful for narrowing large catalogs to relevant tools.
Server Detail, Remote Servers, Clients, and Skills
get_server_detail accepts a slug in owner/repo format (e.g., chroma-core/chroma-mcp) and returns name, description, github_url, stars (may be null due to client-side rendering), and readme_html containing the rendered README content. list_remote_servers provides curated remote MCP servers with OAuth and open-access endpoints, while list_clients covers MCP client applications and list_claude_skills returns reusable agent skills with name, author, description, and url.
The mcpservers API is a managed, monitored endpoint for mcpservers.org — not a raw scraper you maintain. Every endpoint is automatically health-checked on a schedule, and when mcpservers.org changes and a check fails, the API is automatically queued for repair and re-verified. It is built to keep working as the site underneath it changes.
This isn't an official mcpservers.org API — it's an independent, maintained REST wrapper over public data. Where the source has no official API (or only a limited one), Parse gives you a stable contract over a source that never promised one, and keeps it current. Need a new endpoint or field? You can revise it yourself in plain English and the agent rebuilds it against the live site in minutes — contributing the change back to the shared API is free.
Will this API break when the source site changes?+
Is this an official API from the source site?+
Can I fix or extend this API myself if I need a new endpoint or field?+
What happens if I call an endpoint that has an issue?+
- Build a search interface that queries
search_serversby keyword and displays matching MCP servers with descriptions and links. - Populate a categorized MCP server directory using
get_servers_by_categorywith slugs likedatabaseorweb-scraping. - Display a curated 'featured' section in an MCP tooling dashboard using
get_featured_servers. - Retrieve full README documentation for a specific MCP server by passing its GitHub
owner/reposlug toget_server_detail. - List available MCP client applications for a compatibility guide using
list_clients. - Enumerate Claude Skills for an agent-building tool using
list_claude_skillsto surfacename,author, andurl. - Monitor newly added MCP servers by polling
list_serverswithsort=newestand tracking changes over time.
| Tier | Price | Credits/month | Rate limit |
|---|---|---|---|
| Free | $0/mo | 100 | 5 req/min |
| Hobby | $30/mo | 1,000 | 20 req/min |
| Developer | $100/mo | 5,000 | 100 req/min |
One credit = one API call regardless of which marketplace API you call. Exceeding the rate limit returns a 429 response. Authenticate with the X-API-Key header.
Does mcpservers.org have an official developer API?+
What does `get_server_detail` return, and are GitHub stars always available?+
get_server_detail returns name, description, github_url, readme_html, and stars for a server identified by its owner/repo slug. The stars field may be null because the star count on the source site is rendered client-side and is not always retrievable.Can I retrieve a list of all category slugs available on mcpservers.org?+
get_servers_by_category, but there is no endpoint that enumerates all valid category slugs. You would need to supply known slugs such as web-scraping, database, or search. You can fork this API on Parse and revise it to add a category-listing endpoint.Does `search_servers` return a total result count the way `list_servers` does?+
list_servers includes a total_servers field (integer or null). search_servers does not expose a total count — its response contains only page, query, and the servers array. You can fork this API on Parse and revise it to surface a count if the source exposes one.Do any endpoints expose install instructions or configuration schemas for individual MCP servers?+
get_server_detail returns readme_html, which may contain install instructions as unstructured HTML. No endpoint returns parsed configuration schemas or command-line arguments as discrete response fields. You can fork this API on Parse and revise it to extract and structure that content from the README.