M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
A

Apple RAG MCP

Community
Instant Apple docs for AI agents
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #416 of 438
Stars
★ 120 Popular
Transport
Streamable HTTP
Runtime
Hosted (no local runtime)
Credentials
Optional API key
License
MIT
Last commit
Tools
1
29FMRS · D

This MCP server provides powerful Apple documentation search capabilities for AI agents, supporting multiple search modes, and is easy to integrate. It suits scenarios requiring quick access to Apple development resources.

Strongest · Setup experience 8/20 Weakest · Reliability 3/20

Reliability
3/20
Security and permissions
6/20
Maintenance
5/20
Documentation
7/20
Setup experience
8/20
Why each score
Reliability 3/20
Only the supplied server manifest and README were available; no server source, committed tests, or CI workflow files were provided. The manifest is well-formed and declares a streamable HTTP remote (https://mcp.apple-rag.com) with optional auth, and the README claims semantic/keyword/hybrid search, but these are unverified claims. No evidence supports the MCP init/tool-list handshake, actual tool set, error handling, or rate-limit behavior.
Security and permissions 6/20
No red-line issue is triggered: no malware, credential theft, or embedded real tokens are visible because no source was supplied; the only secret reference is an optional Authorization header and no real secret appears in install examples. The main risk is visible but incompletely controlled: this is a third-party remote endpoint, so queries and any token would be sent to an external host. Data retention, processing boundaries, least-privilege behavior, and confirmation mechanisms are not documented, and credential handling cannot be reviewed from source.
Maintenance 5/20
The repository is not archived, has an MIT license, the manifest status is active, and version 4.21.0 is declared; a CI badge is shown but not verifiable from supplied files. There is no commit history, release evidence, dependency-update history, issue-response evidence, or security-response channel in the supplied material. README claims of continuous updates are untrusted, so a moderate-low score is appropriate.
Documentation 7/20
The README gives one-click install buttons for Cursor/VS Code, manual JSON configuration, URL, protocol, optional authentication, feature summaries, and an Agent Skill path; this is usable at a basic level. Deductions: no enumerated MCP tools or parameters, no rate-limit/quota/cost details, no troubleshooting section, and no executable examples or tests to substantiate the claims.
Setup experience 8/20
Setup looks straightforward in the README: one-click installs plus a manual Streamable HTTP JSON example for mainstream clients, and the hosted remote avoids local build/runtime steps. However, these are documentation claims only; there is no verified client-handshake evidence, compatibility matrix, or tests. Under static calibration this cannot receive a higher setup score.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Developers using AI agents for Apple platform development
  • AI coding assistants that need fast and accurate access to Apple docs

Not for

  • Users requiring private documentation or local indexing
  • Users who do not use AI coding tools

Required permissions

  • Network access to connect to the remote server
  • Sending search queries and receiving documentation content

Risks and side effects

  • Dependency on remote service availability
  • Potential for incomplete or outdated information since docs are continuously updated

Setup

Before you start

Runtime：Hosted (no local runtime)

  1. Add the MCP server to a compatible MCP client (e.g., Cursor, Claude Desktop, Cline).
  2. Use the configuration example: {"mcpServers":{"apple-rag-mcp":{"url":"https://mcp.apple-rag.com"}}}.
  3. Alternatively, use the official quick-install buttons for Cursor or VS Code, which configure automatically.
.cursor/mcp.json 
{"mcpServers":{"apple-rag-mcp":{"url":"https://mcp.apple-rag.com"}}}

Shown for Cursor. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "apple-rag-mcp": {
      "url": "https://mcp.apple-rag.com"
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add --transport http apple-rag-mcp https://mcp.apple-rag.com

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

The apple_dev_docs_search tool should appear in your client's tool list; after connecting, a query like 'Search Apple docs for the State property wrapper' should return Apple documentation results.

Troubleshooting

  1. Check network connectivity and remote server status
  2. Ensure MCP client configuration is correct (URL and protocol)
  3. Check authentication token if higher rate limits are needed

Things to try

Once connected, you can ask your AI assistant things like:

  • Search the Apple docs for NavigationStack in SwiftUI
  • Look up the new UIKit APIs in iOS 27
  • Find WWDC26 session transcripts about Swift Concurrency
  • Use hybrid search to find documentation on Combine's debounce operator

Tools 1

apple_dev_docs_search read-only
Searches Apple developer documentation and WWDC transcripts, supporting semantic search, keyword search, and hybrid search, returning relevant documentation snippets and code examples.

Use cases

Quickly find Apple API documentation during AI-assisted coding
Get the latest WWDC26 session content and technical details
Obtain code examples (Swift, Objective-C, SwiftUI) while developing

Supported clients

Cursor
Claude Desktop
Cline

Listed from the project's documentation, not tested by this site.

Overview

Apple RAG MCP is an MCP server that provides AI agents with instant access to Apple developer documentation, offering semantic search, keyword search, and hybrid search, as well as WWDC26 session transcripts. It supports multiple search modes and is designed to be lightweight, providing only necessary information. The content is kept up-to-date with the latest docs, WWDC26 sessions, and Xcode 27 beta references.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 63.3k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 1c9483b0f51c Data synced 2026-10-19 Read the FMRS scoring method