Show HN: AgentSwift – Open-source iOS builder agent

• Notifications You must be signed in to change notification settings
• Fork 1
• Star 20

• Code
• Issues 0
• Pull requests 0
• Actions
• Projects
• Security and quality 0
• Insights

Additional navigation options  mainBranchesTagsGo to fileCodeOpen more actions menu

Folders and files

Name	Name	Last commit message	Last commit date
Latest commit History 27 Commits27 Commits
AgentSwift	AgentSwift
docs	docs
scripts	scripts
.gitignore	.gitignore
AgentSwift-0.1.zip	AgentSwift-0.1.zip
README.md	README.md
agentswift.png	agentswift.png
appcast.xml	appcast.xml
screenshot.png	screenshot.png
screenshot2.png	screenshot2.png
View all files

Repository files navigation

• README

Download AgentSwift-0.1.zip You must install the dependencies listed below in order for the binary to work. See below for setup commands.

Dependencies:

• Xcode
• Xcode command line tools
• xcodebuildmcp
• openspec

A native macOS app that runs an autonomous AI coding agent for Apple platform development. Describe what you want to build, and AgentSwift uses Claude to discover your project, implement changes, build, run, and validate — without you touching Xcode.

AgentSwift drives a multi-phase agentic workflow:

1. Discover — Claude inspects your Xcode project structure and schemes
2. Implement — edits source files to match your request
3. Build — runs xcodebuildmcp to compile
4. Launch / Validate — boots the app on a simulator or macOS, runs UI automation to verify behavior
5. Archive — marks the task complete

• macOS 26.1+
• Xcode
• Node.js / npm
• An Anthropic API key

Install these two CLIs before running the agent:

Provides build, launch, and UI automation capabilities for Xcode projects.

npm install -g xcodebuildmcp

Tracks implementation specs across agent sessions.

npm install -g @fission-ai/openspec

1. Build and run the app in Xcode.
2. Open Settings and enter your Anthropic API key.
3. Select a Project Folder (the root of your Xcode project).
4. Optionally pick an iOS Simulator from the dropdown.
5. Type what you want to build and press Cmd+Return.

On the first run the agent discovers your project's scheme and simulator target. Subsequent runs skip discovery and go straight to implementation.

Model	Use when
Claude Opus 4.7	Complex tasks, large codebases
Claude Sonnet 4.6	Faster iteration, lighter tasks

• Message queuing — if you send a new message while the agent is running, the latest supersedes earlier ones
• Build caching — scheme, project path, and simulator ID are extracted after the first build and reused automatically
• Error escalation — the agent attempts one fix on a failure, then surfaces the error to you rather than looping

AgentSwiftApp.swift    — app entry point
ContentView.swift      — UI, view models, agentic loop
AnthropicService.swift — Anthropic API client (streaming SSE)
ToolExecutor.swift     — bash / read_file / write_file execution
Item.swift             — chat message model

No external Swift dependencies — pure SwiftUI + Foundation.

About

No description, website, or topics provided.

Resources

Readme

Uh oh!

There was an error while loading. Please reload this page.

Activity

Stars

20 stars

Watchers

0 watching

Forks

1 fork Report repository

Releases

No releases published

Packages 0

     

Uh oh!

There was an error while loading. Please reload this page.

Contributors

Uh oh!

There was an error while loading. Please reload this page.

Languages

• Swift 94.0%
• Shell 6.0%