bonesso.io

Giving AI Agents Access to App Store and Google Play

AI agents are getting good at writing code, fixing bugs, and even deploying apps. But when it comes to managing App Store Connect or Google Play, they hit a wall. Both platforms are built around web UIs that require clicking through dozens of screens. No API wrapper, no structured output, no way for an agent to just call a command and get JSON back.

I built StoreOps to bridge that gap. It's an open-source CLI that wraps both store APIs into a single tool designed from the ground up for agent integration.

Why Agents Need This

Think about what an AI agent could do with full store access:

  • Monitor reviews and flag negative ones that need a response
  • Update metadata across 20 localizations without manual copy-pasting
  • Check build status and notify you when a review is approved or rejected
  • Pull analytics and generate daily reports on downloads and revenue
  • Manage TestFlight testers and beta groups automatically
  • Submit builds for review as part of a CI/CD pipeline

All of this is possible today, but only if the agent has a tool that speaks structured data. That's what StoreOps provides.

Agent-Friendly by Design

Every design decision was made with automation in mind:

  • JSON output by default: no parsing HTML or scraping tables
  • No interactive prompts: every parameter is an explicit flag, no TTY required
  • Clean exit codes: 0 for success, non-zero with JSON error details on stderr
  • Deterministic pagination: --limit, --next, and --paginate for predictable data
  • Timeout control: --timeout flag prevents hanging in automated environments

An agent can reliably chain commands together:

APPS=$(storeops apple apps list --output json)
APP_ID=$(echo "$APPS" | jq -r '.[0].id')
storeops apple reviews list --app-id "$APP_ID" --limit 10

Or wrap it as a tool in frameworks like Claude Code, LangChain, or any agent that can shell out.

What It Covers

StoreOps handles the full lifecycle for both Apple and Google:

  • Metadata, keywords, and localized store listings
  • Screenshots and app preview videos
  • Builds, submissions, and review status
  • TestFlight and testing tracks
  • In-app purchases and subscriptions
  • Analytics, sales, and crash data
  • Phased releases and age ratings
  • User reviews with reply support

Getting Started

curl -fsSL https://raw.githubusercontent.com/fbonesso/storeops/main/install.sh | sh

storeops auth init
storeops apple apps list
storeops google reviews list --package com.example.app

Pre-built binaries for macOS, Linux, and Windows. Or build from source with Rust.

The Bigger Picture

The trend is clear: agents are moving beyond code generation into full workflow automation. But they can only automate what they can access programmatically. StoreOps is one piece of that puzzle, giving agents the same store management capabilities that previously required a human clicking through a browser.

It's open source on GitHub. If you're building agent workflows that touch app stores, give it a look.