Shipli

Website: https://prasenjeet-symon.github.io/shipli-ai/ Discord: https://discord.gg/rWctcPsP

Store rejections cost days of development time. Shipli is a CLI tool that audits your Flutter source code against Apple App Store and Google Play guidelines using AI.

Catch missing permissions, policy violations, and compliance issues before you submit.

Features

• Dual Store Support — Audit against Apple App Store and Google Play, together or individually.
• Two Audit Modes — Store compliance, code quality, or both in a single run.
• AI-Powered — Choose between Google Gemini or Anthropic Claude. Bring your own API key.
• Up-to-Date Policies — Ships with the latest Apple and Google Play store policies. Works offline.
• Auto-Detection — Automatically detects project type and target platform from your project structure.
• Zero Setup — No compiled artifacts needed. Point it at your Flutter project and run.
• CI-Ready — Designed for automation. Integrates with any CI/CD pipeline.

Installation

npm install -g @prasenjeet/shipli

Requires Node.js 18 or later.

Usage

Quick Start

# One-time setup — select provider, model, and enter API key
shipli init

# Run full audit (auto-detects platform)
shipli --dir ./

# iOS App Store only
shipli --dir ./ --platform ios --mode store

# Google Play only
shipli --dir ./ --platform android --mode store

# Code quality only
shipli --dir ./ --mode code

Commands

Options

Configuration

Run shipli init to create a .shipli config file, or create one manually:

{
  "provider": "claude",
  "key": "sk-ant-...",
  "model": "claude-sonnet-4-6",
  "platform": "both"
}

The CLI looks for .shipli in two places:

1. Project-level — in the --dir path (highest priority)
2. Global — in your home directory ~/.shipli

Resolution order: CLI flags > project .shipli > global ~/.shipli > env vars > defaults.

Add .shipli to your .gitignore — it contains your API key.

Use shipli config to update settings anytime.

Supported Models

Audit Categories

App Store (iOS)

Play Store (Android)

Code Quality

Package Auditing

MCP Server

Shipli includes a built-in Model Context Protocol (MCP) server, so AI coding assistants like Claude Code, Cursor, and Windsurf can run audits directly inside your editor.

Step 1 — Install globally

npm install -g @prasenjeet/shipli

Verify: shipli --version and which shipli-mcp.

Step 2 — Add the MCP server

Option A: Claude Code CLI (recommended)

# Using Claude as the AI provider
claude mcp add shipli \
  --transport stdio \
  --env SHIPLI_PROVIDER=claude \
  --env SHIPLI_MODEL=claude-haiku-4-5 \
  --env ANTHROPIC_API_KEY=your-api-key \
  -- shipli-mcp

# Using Gemini as the AI provider
claude mcp add shipli \
  --transport stdio \
  --env SHIPLI_PROVIDER=gemini \
  --env SHIPLI_MODEL=gemini-2.5-flash \
  --env GEMINI_API_KEY=your-api-key \
  -- shipli-mcp

Option B: Manual JSON config

Add to ~/.claude/.mcp.json:

{
  "mcpServers": {
    "shipli": {
      "command": "shipli-mcp",
      "env": {
        "SHIPLI_PROVIDER": "claude",
        "SHIPLI_MODEL": "claude-haiku-4-5",
        "ANTHROPIC_API_KEY": "your-api-key"
      }
    }
  }
}

For Cursor/Windsurf, add the same config to your editor's MCP settings.

Step 3 — Restart and verify

Restart Claude Code (or run /mcp in chat) to pick up the new server. You should see shipli listed with its two tools.

Available Tools

Both tools return structured JSON with PASS/WARNING/FAIL scores and specific guideline citations.

Usage

Once configured, ask your AI assistant:

"Run a store audit on /path/to/my-flutter-app"
"Review the code quality of this Flutter project"

The assistant will call the appropriate Shipli MCP tool and return the results inline.

CI Integration

# GitHub Actions
- name: Shipli Audit
  run: npx @prasenjeet/shipli --dir ./ --provider claude --key ${{ secrets.ANTHROPIC_API_KEY }}

The CLI exits with code 1 on failure, making it easy to gate deployments.

Telemetry

Shipli collects anonymous usage metrics to help improve the tool. No project data, file contents, API keys, or personally identifiable information is ever collected.

What's collected: audit mode, platform, provider, model, pass/fail score, duration, token counts, OS, and CLI version.

Opt out using any of these methods:

# Environment variable
export SHIPLI_TELEMETRY=off

# Or use the Do Not Track standard
export DO_NOT_TRACK=1

Or add "telemetry": false to your ~/.shipli config file.

License

MIT