Directory / Search & discovery
PlugMCP directory MCP
Search a directory of 79,021+ MCP servers and agent skills, compare options, and get the install config for your client. Read-only, no account or API key required (an optional key raises rate limits). Endpoint: https://plugmcp.ai/mcp-server · Install: npx -y plugmcp-mcp · Docs · Registry: ai.plugmcp/directory
Live Answered an MCP handshake without credentials and listed 9 tools.
- Category
- Search & discovery
- Transport
http- Source
- https://github.com/plugmcp-ai/plugmcp-mcp
- Endpoint
https://plugmcp.ai/mcp-server- MCP Registry
ai.plugmcp/directory- Reports itself as
- plugmcp-directory 0.3.0
- Protocol
- 2025-06-18
- Tools
- 9
- Context cost
- about 2,256 tokens for the tool definitions (estimated at four characters per token)
- Tool definitions last changed
- no change seen since first check
- Handshake time
- 215 ms
- Repository
- 2 stars · last push 2026-11-14 · MIT
- Last checked
- 2026-11-22 11:27:45 UTC
- Listed since
- 2026-11-18
Health and compatibility
| Endpoint reachable | Verified | Responded in 215 ms |
|---|---|---|
| MCP handshake (initialize) | Verified | Completed without credentials |
| Tool discovery (tools/list) | Verified | 9 tools listed |
| Authentication | Verified | None needed to connect |
| Protocol version | Verified | 2025-06-18 |
| Transport | Verified | Streamable HTTP |
| Capabilities declared | Verified | tools |
Last checked . Verified means the check saw it happen; it says nothing about whether the server is safe.
Recent history
5 daily checks so far; median response 231 ms. Green answered, violet asked for auth, red did not answer.
Set up in your client
Claude Code
Run this in a terminal:
claude mcp add --transport http plugmcp-directory-mcp https://plugmcp.ai/mcp-serverClaude Desktop
Open Settings → Connectors → Add custom connector and paste this URL:
https://plugmcp.ai/mcp-serverOr bridge it through mcp-remote in claude_desktop_config.json (needs Node.js):
{
"mcpServers": {
"plugmcp-directory-mcp": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://plugmcp.ai/mcp-server"
]
}
}
}Cursor
Or add it to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):
{
"mcpServers": {
"plugmcp-directory-mcp": {
"url": "https://plugmcp.ai/mcp-server"
}
}
}VS Code
Or add it to .vscode/mcp.json:
{
"servers": {
"plugmcp-directory-mcp": {
"type": "http",
"url": "https://plugmcp.ai/mcp-server"
}
}
}Or from a terminal:
code --add-mcp '{"name": "plugmcp-directory-mcp", "type": "http", "url": "https://plugmcp.ai/mcp-server"}'Windsurf
Add it to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"plugmcp-directory-mcp": {
"serverUrl": "https://plugmcp.ai/mcp-server"
}
}
}Other clients
Most clients accept this shape. The endpoint is:
https://plugmcp.ai/mcp-server{
"mcpServers": {
"plugmcp-directory-mcp": {
"type": "http",
"url": "https://plugmcp.ai/mcp-server"
}
}
}If it does not connect
- Restart the client after editing a config file, and keep keys out of files you commit.
- If the tools do not appear, open the client's MCP log: a 401 means the credentials were not accepted, and a timeout means the server did not answer.
Tools 9
Names, descriptions and input schemas exactly as the server returned them on 2026-11-22. This is the server author's own text. Nothing is run from this page. Labels such as read-only are declared by the server and are not verified. Raw JSON
compare_serversread-onlyCompare two to five servers side by side by slug: trust signals, auth, hosting, transport, tools count, …
Compare two to five servers side by side by slug: trust signals, auth, hosting, transport, tools count, license.
| Parameter | Type | Description | |
|---|---|---|---|
slugs | string[] | required | Slugs from search results. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "compare_servers",
"arguments": {
"slugs": []
}
}
}get_installread-onlyThe exact config or command to install one server in one client: Cursor, Claude Code, Claude Desktop, VS …
The exact config or command to install one server in one client: Cursor, Claude Code, Claude Desktop, VS Code, Windsurf, ChatGPT, Codex, Gemini CLI, Grok Build, Grok Bot, or plain JSON. For Grok Bot the value is the message to send in its chat. Includes the config file path and a one-click deeplink when the client has one.
| Parameter | Type | Description | |
|---|---|---|---|
slug | string | required | Listing slug from search results, for example github-github-mcp-server. |
client | string | optional | The agent the user installs into. grok is Grok Build (terminal) plus grok.com; grok-bot is Grok Bot, which takes a chat message. json returns a plain mcpServers block for any other client. One of: cursor, claude-code, claude-desktop, vscode, windsurf, chatgpt, codex, gemini-cli … |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_install",
"arguments": {
"slug": "..."
}
}
}get_serverread-onlyDetails for one server by slug: description, tools, auth, transport, health, packages, and links. …
Details for one server by slug: description, tools, auth, transport, health, packages, and links. include_readme adds the README (trimmed).
| Parameter | Type | Description | |
|---|---|---|---|
slug | string | required | Listing slug from search results, for example github-github-mcp-server. |
include_readme | boolean | optional | Also return the README, trimmed to 8,000 characters. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_server",
"arguments": {
"slug": "..."
}
}
}get_skillread-onlyOne skill by slug: what it does, where the SKILL.md lives, and the install command for Cursor, Claude Code, …
One skill by slug: what it does, where the SKILL.md lives, and the install command for Cursor, Claude Code, Codex and Grok Build, plus the chat message that installs it in Grok Bot. include_body adds the SKILL.md text (trimmed).
| Parameter | Type | Description | |
|---|---|---|---|
slug | string | required | Skill slug from search_skills. |
include_body | boolean | optional | Also return the SKILL.md text, trimmed to 6,000 characters. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_skill",
"arguments": {
"slug": "..."
}
}
}list_categoriesread-onlyDirectory categories with listing counts. Use a slug as the category filter in search_servers.
Directory categories with listing counts. Use a slug as the category filter in search_servers.
Takes no parameters.
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "list_categories",
"arguments": {}
}
}recommend_serversread-onlyDescribe what the user wants to do in plain words. Returns the best three servers, each with why it fits, …
Describe what the user wants to do in plain words. Returns the best three servers, each with why it fits, auth, remote or local, health, and the install snippet for the client.
| Parameter | Type | Description | |
|---|---|---|---|
task | string | required | What the agent should be able to do, for example query my Postgres database or send email from Gmail. |
client | string | optional | The agent the user installs into. grok is Grok Build (terminal) plus grok.com; grok-bot is Grok Bot, which takes a chat message. json returns a plain mcpServers block for any other client. One of: cursor, claude-code, claude-desktop, vscode, windsurf, chatgpt, codex, gemini-cli … |
constraints | object | optional | Optional filters. Leave a filter out to allow every server; set it to true to narrow the results. |
limit | integer | optional | How many to return. Default 3. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "recommend_servers",
"arguments": {
"task": "..."
}
}
}search_serversread-onlyFind MCP servers by product name, publisher, or topic. Ranked by relevance (exact name, then publisher, then …
Find MCP servers by product name, publisher, or topic. Ranked by relevance (exact name, then publisher, then tags and description); the product's own server comes first when there is one (jira finds Atlassian's), duplicates merged. Returns compact cards with trust signals; call get_install next. When the user describes a job rather than a product, recommend_servers is one call instead of two.
| Parameter | Type | Description | |
|---|---|---|---|
query | string | optional | Product, publisher, or topic, for example github, postgres, send email. Empty lists the most used servers. |
category | string | optional | Category slug: developer-tools, design-tools, databases, browser-automation, web-scraping, cloud-infrastructure, devops, productivity, communication, knowledge-memory, search, finance, security, data-science-ml, marketing, e-commerce, content-management, … |
client | string | optional | Only servers that work in this client (chatgpt keeps remote servers only). With fields=[install], adds the snippet for it. One of: cursor, claude-code, claude-desktop, vscode, windsurf, chatgpt, codex, gemini-cli … |
voice_ready | boolean | optional | Only servers suited to voice agents. |
remote_only | boolean | optional | Only hosted servers with a remote URL, no local install. |
official_only | boolean | optional | Only servers published by the product they connect to. |
verified_only | boolean | optional | Only servers that passed the live health check. |
limit | integer | optional | Results per page. Default 8. Without a key at most 10. |
cursor | string | optional | next_cursor from the previous page. |
fields | string[] | optional | Extra card fields. install (needs client) and tools_count need a PlugMCP key. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "search_servers",
"arguments": {}
}
}search_skillsread-onlyFind agent skills (SKILL.md packs for Claude Code, Cursor, Codex, Grok Build and Grok Bot) by name, publisher …
Find agent skills (SKILL.md packs for Claude Code, Cursor, Codex, Grok Build and Grok Bot) by name, publisher or topic. Call get_skill for the install command.
| Parameter | Type | Description | |
|---|---|---|---|
query | string | optional | Topic or name, for example pdf, frontend design, or testing. Empty lists the most starred skills. |
category | string | optional | Category slug. |
limit | integer | optional | Results per page. Default 8. Without a key at most 10. |
cursor | string | optional | next_cursor from the previous page. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "search_skills",
"arguments": {}
}
}whats_newread-onlyServers added to the directory recently, best first, plus health changes. Defaults to the last 7 days.
Servers added to the directory recently, best first, plus health changes. Defaults to the last 7 days.
| Parameter | Type | Description | |
|---|---|---|---|
since | string | optional | ISO date or timestamp, for example 2026-11-01. Default: 7 days ago. At most 90 days back. |
category | string | optional | Category slug from list_categories. |
remote_only | boolean | optional | Only hosted servers with a remote URL. |
limit | integer | optional | Results per page. Default 8. Without a key at most 10. |
cursor | string | optional | next_cursor from the previous page. |
response_format | string | optional | json (default) or markdown for a human-readable text block. structuredContent is always JSON. One of: json, markdown |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "whats_new",
"arguments": {}
}
}Badge
Maintain this server? Put its live status in your README:
[](https://mcp.aianytime.net/s/plugmcp-directory-mcp/)Edit this entry · Report a problem
More in Search & discovery
Checks record what a server answered, not whether it is safe. A listing is not a security audit; read a server's source and permissions before you connect it. Add a server · JSON · llms.txt · llms-full.txt
Browse: Agents, IDEs & developer tools · Search & discovery · Knowledge, memory & context · Media & generation · Security · Finance & markets · Marketing, content & social · Cloud, ops & data · Commerce, travel & logistics · Work & productivity · No account needed · Local servers