About App Store Connect MCP
App Store Connect MCP is a server for reading and managing releases, TestFlight, listings, reviews, reports, and related account areas. It defaults to read-only access and requires confirmation for actions that become public or are difficult to undo.
Key features
- 01Manage releases and TestFlight
- 02Work with listings and screenshots
- 03Read reviews and sales reports
- 04Handle IAPs, signing, and Xcode Cloud
- 05Enable optional toolsets
- 06Interactive views for release data
Use cases
- 01Review app feedback and sales
- 02Prepare a release and inspect metadata changes
- 03Monitor post-release health with insights
- 04Manage TestFlight builds and signing
MCP support
Provides App Store Connect capabilities as tools for MCP clients, with toolsets that can be enabled or disabled. It is read-only by default and asks before public or consequential actions.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README
app-store-connect-mcp
MCP server for App Store Connect. MCP is the plug-in format AI clients like Claude, Cursor and Codex use, so with this one your agent can read and manage your App Store Connect account. Read-only by default, and it asks before anything goes public. The binary is app-store-connect-mcp.
Install
Paste this into whatever AI is writing your app right now (we won't tell):
Set up the app-store-connect-mcp server for me: https://github.com/idoo/app-store-connect-mcp/blob/main/docs/INSTALL.md
Or by hand, in three steps:
curl -fsSL https://github.com/idoo/app-store-connect-mcp/releases/latest/download/install.sh | sh # or: brew install idoo/tap/app-store-connect-mcp
app-store-connect-mcp auth add # asks for your API key, tests it with Apple
app-store-connect-mcp config install claude-code # or claude-desktop, cursor, codex, vscode, windsurf
Details in docs/INSTALL.md. Upgrading, key rotation and fixes for common errors: docs/FAQ.md.
Compared
| this one | forgeops | zelentsov | Joshua (archived) | asc | |
|---|---|---|---|---|---|
| MCP | + | + | + | + | − |
| Read-only default | + | − | − | − | − |
| Asks before publishing | + | − | ~ | − | ~ |
| Injection guard | + | − | − | − | − |
| Release preflight | + | − | ~ | − | + |
| Sales reports | + | − | + | + | + |
| One binary, any OS | + | + | − | − | + |
~ partial
Security
It can't break anything until you let it. Anything public, money-moving or hard to undo for your users and team waits for your yes. Build uploads and Xcode Cloud branch runs don't ask; pull request runs do. A 1-star review that says "ignore previous instructions" is still just a 1-star review. More in docs/SAFETY.md.
What it does
Everything you'd click in App Store Connect: releases, TestFlight, listings, screenshots, reviews, sales, IAPs, signing, Xcode Cloud. For everything else there's raw API access. Tool list, config.
After launch it keeps watching, if you turn on the insights toolset: post-release health with a guard that pauses a phased rollout from cron, a "what's new since yesterday" digest, the analytics funnel, and keyword rank tracking. In Claude Desktop, claude.ai, ChatGPT and VS Code the release checklist, metadata diffs, screenshots, reviews and funnel show up as small interactive views.
A toolset is a group of tools you switch on or off, so your agent only sees what you use. The release, TestFlight, metadata, reviews, reports and screenshots ones are on out of the box. Add others with --toolsets +insights,+signing or ASC_MCP_TOOLSETS=+insights, or ask the agent to turn one on mid-chat. The list is in docs/CONFIG.md.
License
MIT © Ivan Verevkin
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.