mcpField
M
MCP Charts

MCP Charts

Remote
mcpcharts.comSearch0Updated 3w ago

About MCP Charts

MCP Charts is a hosted, read-only catalog search service for agents looking for MCP servers, skills, plugins, agents, prompts, and rules. It ranks and searches public resources and returns compatibility evidence and setup links without installing or executing them.

Key features

  1. 01Rank resource recommendations by task
  2. 02Search across six resource types
  3. 03Retrieve compatibility and freshness details
  4. 04Return upstream setup links
  5. 05Hosted access without account or API key

Use cases

  1. 01Find an MCP server for a task
  2. 02Search for skills or plugins by keyword
  3. 03Check a catalog entry's client compatibility
  4. 04Get setup links before installing a resource

MCP support

MCP Charts is itself a hosted MCP server available over Streamable HTTP. Its read-only tools recommend, search, and retrieve catalog entries; they do not install or execute resources.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcpcharts.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

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

3
ParameterTypeRequiredDescription
taskstringYes
typeenumNoskill | mcp | plugin | agent | prompt | rule
limitintegerNo
clientstringNo

Details

From the project README

MCP Charts logo

MCP Charts

The MCP for MCPs. MCP Charts is a hosted, read-only MCP server. Your agent can use it to find the right MCP servers, agent skills, plugins, agents, prompts and rules for a task.

  • Endpoint: https://mcpcharts.com/mcp
  • Transport: Streamable HTTP
  • Auth: none. You don't need an account or API key.
  • Official MCP Registry: com.mcpcharts/mcpcharts

This repository holds connection docs and registry metadata only. The service is hosted, so there is nothing to install or run locally.

Connect

Claude Code

claude mcp add --transport http mcpcharts https://mcpcharts.com/mcp

Cursor, VS Code, Cline, Windsurf and other mcp.json clients

{
  "mcpServers": {
    "mcpcharts": {
      "url": "https://mcpcharts.com/mcp"
    }
  }
}

Some clients use "type": "streamable-http" or "transport": "http" alongside url. Check your client's remote-server docs.

Claude.ai / ChatGPT: add https://mcpcharts.com/mcp as a custom connector.

Tools

All tools are read-only. They never install or execute anything.

ToolWhat it doesInputs
recommend_resourcesGives a ranked shortlist for a task. Keyword relevance comes before repository stars.task, optional client, type, limit (1–8)
search_resourcesSearches keywords across all resource types, up to 24 results per page.query, optional type, page
get_resourceReturns compatibility evidence, freshness and the upstream setup link for one catalog entry.id

type is one of skill, mcp, plugin, agent, prompt or rule.

Example prompts
  • "Find me an MCP server for querying Postgres from Claude Code."
  • "What skills exist for writing Playwright tests?"
  • "Recommend a plugin for reviewing pull requests in Cursor."

How the catalog works

The catalog is refreshed continuously from public sources, including the official MCP Registry. Repository stars and activity describe repositories, not individual resources. Client compatibility comes from manifests and registry declarations; MCP Charts doesn't run runtime tests. Check requirements such as "free" or "local-only" upstream before you install anything.

Requests are rate-limited and results are bounded.

Links


Copyright © 2026 MCP Charts. All rights reserved. This repository contains documentation only. No license to the MCP Charts service, data or software is granted.

Comments

Markdown supported

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.