About Socketlist
Socketlist is a free, read-only remote MCP server for searching an independent directory of more than 800 open-source MCP servers. It returns server details and known install configurations based on public GitHub data.
Key features
- 01Search servers by query and category
- 02Retrieve server details and install configs
- 03Browse categories and counts
- 04Find curated servers for a use case
Use cases
- 01Find an MCP server for a particular task
- 02Compare directory entries and server details
- 03Retrieve a known installation configuration
MCP support
Exposes read-only directory search and lookup tools to MCP clients over stateless Streamable HTTP, without authentication.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://socketlist.pages.dev/mcpclaude mcp add --transport http thunderbid-socketlist https://socketlist.pages.dev/mcp{"mcpServers":{"thunderbid-socketlist":{"type":"http","url":"https://socketlist.pages.dev/mcp"}}}Open in Cursor{"servers":{"thunderbid-socketlist":{"type":"http","url":"https://socketlist.pages.dev/mcp"}}}Open in VS Code{"mcpServers":{"thunderbid-socketlist":{"type":"http","url":"https://socketlist.pages.dev/mcp"}}}{"mcpServers":{"thunderbid-socketlist":{"serverUrl":"https://socketlist.pages.dev/mcp"}}}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.
Tools
4| Parameter | Type | Required | Description |
|---|---|---|---|
| limit | integer | No | |
| query | string | Yes | Keywords, e.g. "postgres", "browser automation", "jira" |
| category | string | No | Optional category slug from list_categories, e.g. "databases" |
Details
From the project READMESocketlist MCP server
A free, read-only remote MCP server for searching Socketlist, an independent directory of 800+ open-source Model Context Protocol servers built from public GitHub data.
- Endpoint:
https://socketlist.pages.dev/mcp - Transport: Streamable HTTP (stateless; JSON responses, no SSE stream)
- Auth: none. No API key, no sign-in.
- Docs: https://socketlist.pages.dev/mcp-server/
Tools
| Tool | Arguments | Returns |
|---|---|---|
search_servers | query, category?, limit? (1–50, default 10) | Matching servers ranked by relevance, then stars |
get_server | slug | Details, README-derived features/tools/requirements, and the install config when one is known |
list_categories | – | All categories with slugs and counts |
best_for | use_case (e.g. claude-code, jira, postgresql) | The curated “best MCP servers for X” list |
All tools are read-only (readOnlyHint: true).
Install
Claude Code
claude mcp add --transport http socketlist https://socketlist.pages.dev/mcp
Claude (desktop, web, mobile): Customize → Connectors → + → Add custom connector, URL https://socketlist.pages.dev/mcp, no sign-in.
Cursor (~/.cursor/mcp.json)
{ "mcpServers": { "socketlist": { "url": "https://socketlist.pages.dev/mcp" } } }
VS Code (.mcp.json or ~/.copilot/mcp-config.json)
{ "mcpServers": { "socketlist": { "type": "http", "url": "https://socketlist.pages.dev/mcp" } } }
Codex (~/.codex/config.toml)
[mcp_servers.socketlist]
url = "https://socketlist.pages.dev/mcp"
Test with the MCP Inspector
npx @modelcontextprotocol/inspector --cli https://socketlist.pages.dev/mcp --transport http --method tools/list
How it works
src/mcp.js is a Cloudflare Pages Function that answers JSON-RPC (initialize, ping, tools/list, tools/call) and reads a static JSON snapshot of the directory that is rebuilt on each refresh. It does not store prompts or tool arguments. There is light, best-effort rate limiting per IP, and CORS is open.
Data comes from public GitHub repositories and READMEs. Install configs are copied from READMEs, not verified: review any server before installing it (security checklist).
Contact
Corrections or removal requests: [email protected], or open an issue.
License
MIT
Comments
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.