asc-feedback-mcp

asc-feedback-mcp

MCP server that exposes App Store Connect TestFlight beta feedback including screenshot submissions, crash submissions, crash logs, and screenshot downloads as tools for MCP-capable clients.

Category
Visit Server

README

asc-feedback-mcp

An MCP (Model Context Protocol) server that exposes App Store Connect TestFlight beta feedback — screenshot submissions, crash submissions, crash logs, and screenshot downloads — as tools for MCP-capable clients. It authenticates to the App Store Connect API with a short-lived ES256 JWT minted from your .p8 API key, so an assistant can pull and triage beta-tester feedback without you ever pasting raw credentials into a conversation. The server runs in Docker and speaks streamable-HTTP MCP at http://127.0.0.1:8081/mcp.

Features

  • 4 tools available for listing and inspecting TestFlight beta feedback and downloading feedback screenshots.
  • Streamable-HTTP MCP transport, served at http://127.0.0.1:8081/mcp.
  • Docker-based deployment via docker compose — no local Python environment needed.
  • Binds to 127.0.0.1 only; the service is unauthenticated and must never be exposed beyond the local machine.

Architecture

classDiagram
    class Config {
        +str key_id
        +str issuer_id
        +str private_key
    }
    class AscClient {
        -str _token
        -float _token_exp
        -_bearer_token() str
        +list_screenshot_feedback(asc_app_id, limit) dict
        +list_crash_feedback(asc_app_id, limit) dict
        +get_crash_log(submission_id) dict
    }
    class Server {
        +list_screenshot_feedback(asc_app_id, limit) dict
        +list_crash_feedback(asc_app_id, limit) dict
        +get_crash_log(submission_id) dict
        +download_screenshot(url) Image
    }
    Server --> AscClient : calls
    AscClient --> Config : reads credentials
sequenceDiagram
    participant Client as MCP Client
    participant Server as FastMCP server
    participant ASCClient as AscClient
    participant ASC as App Store Connect API

    Client->>Server: call tool (e.g. list_crash_feedback)
    Server->>Server: validate parameters
    Server->>ASCClient: forward request
    ASCClient->>ASCClient: reuse cached ES256 JWT (mint only if near expiry)
    ASCClient->>ASC: GET request with Bearer JWT
    ASC-->>ASCClient: JSON response
    ASCClient-->>Server: parsed result
    Server-->>Client: tool result

Usage

The server speaks MCP over streamable-http at:

http://127.0.0.1:8081/mcp

Tools

Tool Parameters Description
list_screenshot_feedback asc_app_id: str, limit: int = 50 List the most recent TestFlight screenshot feedback submissions for an app. Returns raw App Store Connect API JSON. asc_app_id is the app's numeric Apple ID.
list_crash_feedback asc_app_id: str, limit: int = 50 List the most recent TestFlight crash feedback submissions for an app. Returns raw App Store Connect API JSON. asc_app_id is the app's numeric Apple ID.
get_crash_log submission_id: str Fetch the crash log for one crash feedback submission. Returns raw App Store Connect API JSON (.data.attributes.logText holds the log).
download_screenshot url: str Download a TestFlight feedback screenshot and return it as image content. Host validation is best-effort: requires https and an Apple-owned hostname.

asc_app_id must contain digits only, and limit must be between 1 and 200 inclusive. submission_id may only contain letters, digits, and dashes.

Example MCP client config

{
  "mcpServers": {
    "asc-feedback": {
      "type": "streamable-http",
      "url": "http://127.0.0.1:8081/mcp"
    }
  }
}

Or with Claude Code:

claude mcp add --transport http asc-feedback http://127.0.0.1:8081/mcp

The host port (8081) is configurable in docker-compose.yaml, but the bind address must always keep the 127.0.0.1: prefix — never change it to 0.0.0.0 or otherwise expose it publicly.

Getting Started

Prerequisites

  • Docker Desktop or Docker Engine with Docker Compose.
  • An App Store Connect API key.

Get an App Store Connect API key

In App Store Connect, go to Users and Access → Integrations → App Store Connect API (Team Keys) and generate a key there. This gives you the Issuer ID (shown at the top of that page), the Key ID (in the keys table), and a downloadable .p8 private key file (e.g. AuthKey_ABC123DEFG.p8) — save this file, as Apple only lets you download it once.

Value Where it comes from
ASC_KEY_ID Users and Access → Integrations → App Store Connect API, keys table
ASC_ISSUER_ID Users and Access → Integrations → App Store Connect API, shown at the top of the page
ASC_PRIVATE_KEY_B64 Base64 of the .p8 file downloaded when creating the key above
asc_app_id (tool parameter) Apps → (your app) → App Information → General Information → Apple ID

Steps

  1. Clone this repository.

  2. Copy the environment template:

    cp .env.example .env
    
  3. Edit .env and fill in:

    • ASC_KEY_ID — the API key id from Integrations.

    • ASC_ISSUER_ID — the issuer id from Integrations.

    • ASC_PRIVATE_KEY_B64 — the base64-encoded contents of your .p8 file, produced with:

      base64 -i AuthKey_XXXX.p8 | tr -d '\n'
      
  4. Build and start the server:

    docker compose up -d --build
    
  5. Verify it's running by initializing an MCP session against the endpoint:

    curl -i http://127.0.0.1:8081/mcp \
      -H "Content-Type: application/json" \
      -H "Accept: application/json, text/event-stream" \
      -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"verify","version":"1.0"}}}'
    

    Alternatively, add the server to an MCP client using the example config above, and confirm all 4 tools (list_screenshot_feedback, list_crash_feedback, get_crash_log, download_screenshot) appear.

Security

This service implements no authentication. It binds to 127.0.0.1 only and must never be exposed on 0.0.0.0 or any public/non-loopback interface — anyone who can reach the port can pull your beta feedback using your App Store Connect credentials.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured