# Ansight

> Ansight is an SDK and desktop studio that gives AI coding agents runtime visibility into mobile apps. It bridges the gap between the agent's ability to write code and its inability to see what the app actually does at runtime.

## What it does

Ansight has two parts:

1. **Ansight SDK** — a lightweight library that runs inside a mobile app and continuously captures runtime telemetry: logs, exceptions, performance metrics (memory, FPS), lifecycle events, navigation state, and screenshots.

2. **Ansight Studio** — a desktop application that receives telemetry from paired devices over the local network and exposes it to AI agents through MCP (Model Context Protocol).

When an AI agent (Codex, Claude Code, Cursor, or any MCP client) connects to Studio, it can:
- Take screenshots of the running app
- Read runtime logs and exceptions
- Query memory, FPS, and performance metrics
- Inspect the view hierarchy
- Call app-defined tools
- Download app artifacts (databases, files)
- Review session history and saved analyses

All data stays local. Studio runs on the developer's machine, communicates with the device over LAN, and never sends data to any cloud service.

## Supported frameworks

- .NET MAUI — In Beta (iOS, Android, MacCatalyst)
- Flutter — Planned
- React Native — Planned
- iOS (SwiftUI, UIKit) — Planned
- Android (Jetpack Compose, Android Views) — Planned

## How to use with an AI agent

1. Install the Ansight SDK in your mobile app
2. Open Ansight Studio on your desktop
3. Pair your device with Studio (QR code or JSON config)
4. Configure your AI agent to connect to Studio via MCP
5. Ask your agent to use Ansight tools to inspect, debug, or test the app

## Agent resources

Fetch the index that matches what you need. Each lists titles, descriptions, and direct URLs to raw markdown files.

- **Skills** (agent-ready instruction files): https://ansight.ai/skills/index.json
- **Documentation** (SDK, Studio, MCP setup): https://ansight.ai/docs/index.json

## Links

- Website: https://ansight.ai
- Pricing: https://ansight.ai/pricing
- QA Workflow: https://ansight.ai/tester-workflow
- FAQ: https://ansight.ai/faq