# Iterating With AI And MCP

Use this skill to drive Compose HotSwan from an AI agent (Claude Code, Cursor, any MCP...

- Repository: https://github.com/skydoves/compose-performance-skills (524 stars)
- Category: Design
- License: Apache-2.0
- Last updated: 2026-06-16
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add skydoves/compose-performance-skills --skill iterating-with-ai-and-mcp -g -a claude-code -y`
Codex: `npx skills add skydoves/compose-performance-skills --skill iterating-with-ai-and-mcp -g -a codex -y`
Cursor: `npx skills add skydoves/compose-performance-skills --skill iterating-with-ai-and-mcp -g -a cursor -y`
Gemini CLI: `npx skills add skydoves/compose-performance-skills --skill iterating-with-ai-and-mcp -g -a gemini-cli -y`

## Description

Use this skill to drive Compose HotSwan from an AI agent (Claude Code, Cursor, any MCP client) so the agent can edit a Kotlin file, trigger a hot reload, capture a device screenshot, evaluate the result against a design intent, and iterate without a human in the loop. Covers the seven HotSwan MCP tools (hotswan_get_status, hotswan_reload, hotswan_take_screenshot, hotswan_start_snapshot, hotswan_stop_snapshot, hotswan_select_variant, hotswan_build_and_install), the canonical edit-reload-screenshot loop, snapshot-based rollback, and when to fall back to a full install for schema changes. Use when the developer says "get the AI to tune this screen until it matches a mock", asks "can the AI see what changed?" or "can the AI screenshot the device?", sets up a Claude Code or Cursor workflow that needs MCP tool access, or wants AI-driven UI iteration.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Iterating with AI and MCP: let the agent edit, reload, screenshot, and iterate

Compose HotSwan ships an embedded HTTP MCP server inside the IntelliJ plugin. Any MCP-compatible AI client (Claude Code, Cursor, any tool that speaks Model Context Protocol) can call its tools to drive the iteration loop. The agent edits a Kotlin file, calls `hotswan_reload`, captures a screenshot of the running device, evaluates the result against the design intent, and decides the next change. Cycle time is comparable to the human loop, a few seconds per iteration, so the agent can converge on a UI tweak without a human steering each step.

This skill teaches the canonical agent loop, the seven MCP tools (verbatim names), and the safety habits (status check first, snapshot wrapping for rollback, fallback to full install for schema changes) that keep the loop reliable.

## When to use this skill

- The developer wants AI-driven UI iteration: "tune this screen until it matches the mock", "let the agent pick a colour".
- The developer wants the AI to verify its own edits visually instead of guessing whether the change worked.
- The developer is wiring up a Claude Code or Cursor workflow that needs MCP tool access to a running app.
- The user asks "can the AI see what changed?", "can the AI screenshot the device?", "how does Claude Code drive HotSwan?".
- The user mentions "MCP server", "hotswan_reload", "hotswan_take_screenshot", "agent loop", or "snapshot rollback".

## When NOT to use this skill

The rest: https://github.com/skydoves/compose-performance-skills/blob/HEAD/hot-reload/iterating-with-ai-and-mcp/SKILL.md