Загрузка видео...

Не удалось загрузить видео

На главную

XcodeBuildMCP 2.6 is here, and the UI automation tools got a serious glow-up. 🚀 Same task, same simulator, side by side: ⏱ 70% faster (6m57s → 2m06s) 🧠 68% fewer tokens (232.8k → 75k) 🔧 76% fewer tool calls (86 → 21) Watch them race 👇

camsoft2000's profile picture

camsoft2000

4,068 subscribers

28,580 просмотров • 5 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 7

Фото профиля camsoft2000
camsoft20005 месяцев назад

Get it here:

Фото профиля Ewan Makepeace
Ewan Makepeace5 месяцев назад

This looks neat but I am confused. Codex is already driving my XCode by CLI - what does this add that would make my life better?

Фото профиля camsoft2000
camsoft20005 месяцев назад

XcodeBuildMCP proivdes stable agentic workflows for AI agents with highly optimised responses that results in your agent using less tokens and less guessing. Yes agents can build, install etc but they have to discover the project settings, choose the correct simulator, use multiple commands to build, install and launch apps where XcodeBuildMCP provides a single tool to “run” the app reducing latency by reducing number of AI tool call round trips. The main idea is to provide a stable tool surface for the agent. One thing agents can’t do without a tool like this is ui-automation like in the demo. The agent is doing everything, all the simulator interactions are performed by the agent so it can verify its own work. Agent let debugging is easier because each build/run tool automatically starts capturing logs so the agent can tail them in a token efficient manner. Xcodebuild isn’t optimised for agentic use, XcodeBuildMCP is.

Фото профиля netox
netox5 месяцев назад

bro, I am so late for the boat, it save me and my time. Sincerely thanks!

Фото профиля Nenad
Nenad5 месяцев назад

good job i use it on every projects

Фото профиля HelloShit
HelloShit5 месяцев назад

@PaulSolt Could it be able to use for testing a macOS app?

Фото профиля camsoft2000
camsoft20005 месяцев назад

@PaulSolt So this is just Simulator automation which is hard, macOS is much easier as there are public APIs for that. There are already tools available to controlling macOS apps.

Похожие видео

2/It's a step up on reasoning and coding. compared to Muse Spark 1.2, Muse Spark 1.3 wastes fewer turns, uses ~20% fewer tool calls, ~25% fewer tokens, and holds onto requirements well during long-horizon tasks.
1:14

2/It's a step up on reasoning and coding. compared to Muse Spark 1.2, Muse Spark 1.3 wastes fewer turns, uses ~20% fewer tool calls, ~25% fewer tokens, and holds onto requirements well during long-horizon tasks.

Alexandr Wang

35,480 просмотров • 1 месяц назад

Still not convinced of the benefits of UI Automation with XcodeBuildMCP? How about having your very own QA tester? The prompt "Please go ahead and test different calculations":
3:25

Still not convinced of the benefits of UI Automation with XcodeBuildMCP? How about having your very own QA tester? The prompt "Please go ahead and test different calculations":

camsoft2000

21,685 просмотров • 1 год назад

I gave both Claude Code and Mentat the same task and had them race, side by side see below for the highlights (and the unexpected twist at 0:33), full video in next tweet
2:13

I gave both Claude Code and Mentat the same task and had them race, side by side see below for the highlights (and the unexpected twist at 0:33), full video in next tweet

Scott Swingle

48,124 просмотров • 1 год назад

Claude Code just got a memory upgrade. And it’s kind of insane. A free tool called Claude Mem lets Claude remember everything across sessions. The result? Up to 95% fewer tokens used. Which means faster workflows and way cheaper AI automation. This fixes Claude Code’s biggest weakness.
0:24

Claude Code just got a memory upgrade. And it’s kind of insane. A free tool called Claude Mem lets Claude remember everything across sessions. The result? Up to 95% fewer tokens used. Which means faster workflows and way cheaper AI automation. This fixes Claude Code’s biggest weakness.

Julian Goldie SEO

11,989 просмотров • 7 месяцев назад

Coming soon: XcodeBuildMCP seamlessly integrates with the Xcode IDE! Its tools automatically use the Xcode IDE's project, scheme and simulator selections, extending Xcode’s AI coding agents with new abilities: run on device/simulator, tap, swipe, type, capture screenshots/video, debug, log, and more. They complement Xcode’s built-in toolset. We don’t duplicate what Xcode already does well (like build/test), we add the capabilities that unlock new workflows. And that's not all. If you're using XcodeBuildMCP with your own coding agent, and you have the latest Xcode open, XcodeBuildMCP will proxy Xcode MCP's tools, so you don't need to set up another MCP server. It just works.
1:39

Coming soon: XcodeBuildMCP seamlessly integrates with the Xcode IDE! Its tools automatically use the Xcode IDE's project, scheme and simulator selections, extending Xcode’s AI coding agents with new abilities: run on device/simulator, tap, swipe, type, capture screenshots/video, debug, log, and more. They complement Xcode’s built-in toolset. We don’t duplicate what Xcode already does well (like build/test), we add the capabilities that unlock new workflows. And that's not all. If you're using XcodeBuildMCP with your own coding agent, and you have the latest Xcode open, XcodeBuildMCP will proxy Xcode MCP's tools, so you don't need to set up another MCP server. It just works.

camsoft2000

14,608 просмотров • 8 месяцев назад

Sonnet 5.5 is priced the same as Sonnet 5, but it typically needs far fewer tokens to do the same work. In our testing, it costs up to 30% less per task than its predecessor. It also generates output more than 30% faster, making it our fastest Sonnet model to date.
0:49

Sonnet 5.5 is priced the same as Sonnet 5, but it typically needs far fewer tokens to do the same work. In our testing, it costs up to 30% less per task than its predecessor. It also generates output more than 30% faster, making it our fastest Sonnet model to date.

Claude

413,753 просмотров • 1 месяц назад

I ran the same task on Claude Code and DeepSeek's new agent harness. One cost $150. The other cost $2. Today we're launching (AgentSky), the "OpenRouter for Agents" — one API → Claude Code, Codex, DeepSeek, Kimi, OpenCode, and every major agent in the cloud. And Agent Playground on top: race them on your own task, with your real tools (GitHub, Gmail, more), side by side in a browser: time, cost, tokens burnt. Guess which one was $2.
1:07

I ran the same task on Claude Code and DeepSeek's new agent harness. One cost $150. The other cost $2. Today we're launching (AgentSky), the "OpenRouter for Agents" — one API → Claude Code, Codex, DeepSeek, Kimi, OpenCode, and every major agent in the cloud. And Agent Playground on top: race them on your own task, with your real tools (GitHub, Gmail, more), side by side in a browser: time, cost, tokens burnt. Guess which one was $2.

Xiaoyin Qu

2,227,402 просмотров • 2 месяцев назад

"You can't just turn round and punch someone" - Kris Boyd Watch them side by side. Identical. Arm raised at the same time, strike opponent at the same time. Same force. Same action. Same intent. Different outcome.
0:21

"You can't just turn round and punch someone" - Kris Boyd Watch them side by side. Identical. Arm raised at the same time, strike opponent at the same time. Same force. Same action. Same intent. Different outcome.

Celtic Thoughts

206,549 просмотров • 8 месяцев назад

to the people asking for the “old bts” back..literally watch danger and 2.0 MV side by side. SAME VIBE, SAME ENERGY, NO ROOTS LOST HERE
0:34

to the people asking for the “old bts” back..literally watch danger and 2.0 MV side by side. SAME VIBE, SAME ENERGY, NO ROOTS LOST HERE

chan⁷ will see BTS

228,459 просмотров • 7 месяцев назад

Stop letting your AI agent guess what’s inside the iOS Simulator. I built RocketSim’s CLI + Agent Skill so agents can inspect screens, tap, type, recover from state changes, and even validate accessibility support with far fewer tokens. Benchmarked against a popular alternative: 19% faster, 63% fewer estimated tokens, 0 wrong taps. Demo below 👇
5:35

Stop letting your AI agent guess what’s inside the iOS Simulator. I built RocketSim’s CLI + Agent Skill so agents can inspect screens, tap, type, recover from state changes, and even validate accessibility support with far fewer tokens. Benchmarked against a popular alternative: 19% faster, 63% fewer estimated tokens, 0 wrong taps. Demo below 👇

Antoine v.d. SwiftLee 

29,098 просмотров • 5 месяцев назад

At Perplexity, GPT-5.5 in Codex helped build an internal tool in under an hour. In Perplexity Computer workflows, GPT-5.5 used 56% fewer tokens on the same complex tasks, creating faster feedback loops for users.
1:05

At Perplexity, GPT-5.5 in Codex helped build an internal tool in under an hour. In Perplexity Computer workflows, GPT-5.5 used 56% fewer tokens on the same complex tasks, creating faster feedback loops for users.

OpenAI Developers

68,320 просмотров • 6 месяцев назад

Jokic mentioned that the same AG dunk happened in the 2010 Euroleague Final 4 Semifinal between Olympiakos and Partizan. 2.6 sec left vs a buzzer beater. But here they are side by side. #Nuggets
0:31

Jokic mentioned that the same AG dunk happened in the 2010 Euroleague Final 4 Semifinal between Olympiakos and Partizan. 2.6 sec left vs a buzzer beater. But here they are side by side. #Nuggets

will jones ⚒️

270,571 просмотров • 1 год назад

Claude code 4.7 v.s. deepseek v4+ hermes. Same coding task side by side comparison.
5:05

Claude code 4.7 v.s. deepseek v4+ hermes. Same coding task side by side comparison.

Xiaoyin Qu

90,634 просмотров • 6 месяцев назад

Grok 4.6 is really efficient.....it uses fewer steps and tokens, completes tasks faster and cheaper
0:32

Grok 4.6 is really efficient.....it uses fewer steps and tokens, completes tasks faster and cheaper

X Freeze

36,885 просмотров • 2 месяцев назад

We've been testing Gemini 3 for the past few weeks, and here's what stands out: It's faster, the tool calling is sharper, and the UI taste is finally dialed in. In this test, we gave Gemini 3 (left side) and Gemini 2.5 (right side) the same tasks and see the difference in the results.
0:43

We've been testing Gemini 3 for the past few weeks, and here's what stands out: It's faster, the tool calling is sharper, and the UI taste is finally dialed in. In this test, we gave Gemini 3 (left side) and Gemini 2.5 (right side) the same tasks and see the difference in the results.

Cline

31,469 просмотров • 11 месяцев назад

Side by side. Same as last time, a few of these are not the same shot cause I couldn't find them.
1:51

Side by side. Same as last time, a few of these are not the same shot cause I couldn't find them.

Kurapees

456,731 просмотров • 1 месяц назад

🔥 Masterclass here from the Vice President on dealing with the Left’s last refuge: THE RACE CARD. And by way, YES, making neighborhoods and cities safer is saving Black Lives. I thought they mattered?… Americans of every race want the same thing: safe streets, good schools, and a chance to build a better life. The race-hustling act is getting old, and fewer and fewer people are buying tickets. This is NOT going the way “The View”hoped.
1:42

🔥 Masterclass here from the Vice President on dealing with the Left’s last refuge: THE RACE CARD. And by way, YES, making neighborhoods and cities safer is saving Black Lives. I thought they mattered?… Americans of every race want the same thing: safe streets, good schools, and a chance to build a better life. The race-hustling act is getting old, and fewer and fewer people are buying tickets. This is NOT going the way “The View”hoped.

Wesley Hunt

164,799 просмотров • 4 месяцев назад

Meghan said she just wanted to “get to the other side of this.” Six years later and the GPS apparently said: “Make a U-turn when possible.” 🇬🇧😭 Sometimes “the other side” is just the same place with fewer benefits. #MeghanMarkleExposed #MeghanMarkleIsANarcissist #RevealingTheNarc
0:24

Meghan said she just wanted to “get to the other side of this.” Six years later and the GPS apparently said: “Make a U-turn when possible.” 🇬🇧😭 Sometimes “the other side” is just the same place with fewer benefits. #MeghanMarkleExposed #MeghanMarkleIsANarcissist #RevealingTheNarc

Revealing

44,125 просмотров • 2 месяцев назад

Turborepo 2.9 is here! → Up to 96% faster → 𝚝𝚞𝚛𝚋𝚘 𝚚𝚞𝚎𝚛𝚢, Native OpenTelemetry, structured logging, more... → 70% fewer known Issues compared to 2.8
10:58

Turborepo 2.9 is here! → Up to 96% faster → 𝚝𝚞𝚛𝚋𝚘 𝚚𝚞𝚎𝚛𝚢, Native OpenTelemetry, structured logging, more... → 70% fewer known Issues compared to 2.8

Anthony Shew

49,431 просмотров • 7 месяцев назад

The same 40 foot putt in the same group. The player using AimPoint made 77 fewer footprints and finished his read 2.5 times faster. Don’t believe the narrative.
1:41

The same 40 foot putt in the same group. The player using AimPoint made 77 fewer footprints and finished his read 2.5 times faster. Don’t believe the narrative.

AimPoint Golf

95,052 просмотров • 9 месяцев назад