Interceptor Ios logo

Interceptor Ios

Organization
Hacker-Valley-Media
interceptor-ios

Drive any installed app on an owned, unlocked, Developer-Mode iPhone via interceptor ios *: ref-tagged element trees, deterministic coordinate taps (click), reliable text entry (type/keys), scroll, drag, multi-touch gestures (gesture), hardware buttons (press), screenshots, a continuous JPEG frame stream (stream/frame), installed-app listing, and app launch/activate/terminate. The phone runs an on-device XCUITest runner (InterceptorRunner) that dials into the daemon over WiFi — no cable once paired, no WebDriverAgent. Address a phone with --on <name> or ios:<udid>; phones auto-connect on the first verb. Use for iPhone app automation. Not for the iOS Simulator UI of a Mac app, and not for content inside a browser tab (use interceptor-browser) or a macOS app (use interceptor-macos).

Overview

PublisherHacker-Valley-Media
RepositoryInterceptor
Skill nameinterceptor-ios
Stars
526
Forks
76
Bundled files
2
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Hacker-Valley-Media on GitHub. Read the source before you install it.

Installation

Install the Interceptor Ios AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Hacker-Valley-Media/Interceptor/tree/main/.agents/skills/interceptor-ios
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Hacker-Valley-Media/Interceptor.git /tmp/Interceptor
mkdir -p .claude/skills
cp -r /tmp/Interceptor/.agents/skills/interceptor-ios .claude/skills/interceptor-ios
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Interceptor Ios in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Interceptor Ios on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Interceptor Ios is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Interceptor iOS

Agent-operator skill for the iOS surface of Interceptor. Use the interceptor ios * CLI to drive any installed app on an owned, unlocked, Developer-Mode iPhone: ref-tagged accessibility trees, deterministic coordinate taps, reliable text entry, scroll/drag, hardware buttons, screenshots, and app lifecycle.

The phone runs Interceptor's own on-device XCUITest runner (InterceptorRunner — not WebDriverAgent). The runner dials into the daemon over a WebSocket, exactly like the browser extension; the daemon drives it from there. Once the phone is paired over WiFi, no cable is needed. For content inside a browser tab load interceptor-browser; for a macOS app load interceptor-macos.

This installed skill is self-contained. Source checkouts also have AGENTS.md, but packaged users may only have the skill directory below /Library/Application Support/Interceptor/skills.

Fast Path

bash
interceptor ios devices                        # 1. Phones with the agent installed (+ aliases)
interceptor ios status                          # 2. Per-phone connection state
interceptor ios tree --on phone                 # 3. Auto-connects, returns the ref-tagged element tree
interceptor ios find --label "Slack" --on phone # 4. Locate an element by label/role
interceptor ios click e29 --on phone            # 5. Deterministic coordinate tap at the ref's frame center
interceptor ios type e243 "hello" --on phone    # 6. Focus the field, then type
interceptor ios screenshot --on phone           # 7. Capture the screen (VLM-budget resized)

Omit --on <name> when only one phone is set up — it's used by default. Set an alias once with interceptor ios name <udid> phone, then always use --on phone.

Treat eN refs as short-lived. The UI changes between calls; re-read with interceptor ios tree before acting. Refs carry frames and resolve to coordinates, so a tap is deterministic even if the underlying element handle went stale.

Apps with no usable tree (games, canvases): click, drag, and scroll also take screen points, drag X,Y X,Y --duration 2 is a long press, and interceptor macos vision text --image <screenshot> reads the screen. For a loop that needs a fresh look every few hundred milliseconds, ios stream start --out <path> has the runner push JPEG frames over Wi-Fi and rewrite that one file, and ios gesture presses with several fingers at once (a held pedal plus a tapped arrow). The recipes and the box-to-point conversion are in the command catalog.

The Model

  • The device dials in. Ordinary runner verbs auto-launch on an unlocked phone and connect back over the network. The daemon hands the runner a VPN (Tailscale) address when the Mac has one, because iOS silently denies a backgrounded runner's LAN connection until the user has granted it Local Network access (Settings › Privacy & Security › Local Network), and a VPN address is exempt from that check; interceptor ios status shows the address as dialBack / dialBackVia. Put the phone on the same VPN, or grant that switch once and LAN dial-back works too. Unlock and its probe require an already connected resident runner and do not auto-launch.
  • Unlocked + foreground matters. A locked phone refuses app launches. While the runner is connected and resident, interceptor ios unlock --secret ios-passcode attempts vault-backed passcode entry; success requires an observed unlocked state. ios unlock --probe checks without typing and can succeed while reporting locked: true. Both commands fail immediately when disconnected, without trying to launch XCTest on a locked phone. Unlock once and run ios tree to connect the runner first. Keep Auto-Lock off while working.
  • Passcodes come from the vault, never from chat. Nothing can fake Face ID or Apple Pay. On a Face ID sheet tap "Enter Passcode", then interceptor ios type <ref> --secret ios-passcode (or ios keys --secret ios-passcode); the runner types into SpringBoard when the sheet owns the keyboard. Register the passcode once with interceptor macos secret register ios-passcode --target ios.
  • The XCTest authorization sheet is a human gate. Stop and ask. The first XCUITest launch after a reboot, and any later re-authorization, shows "Enter iPhone Passcode for XCTest — Enable UI Automation" on the phone; a runner verb then fails with Not authorized for performing UI testing actions or times out on XCTestManager_IDEInterface. Nothing on the Mac can enter it: the runner is the process it blocks, so ios unlock and keys --secret cannot help; AccessibilityAudit and Accessibility Inspector can read the sheet but every action on it reports unsupported and leaves the field at 0 of 6; Switch Control cannot target a digit; iPhone Mirroring does not forward keystrokes to it; re-signing Apple's tools strips the private entitlements they need. When you see the sheet, report it and ask for the passcode to be tapped on the phone (or typed by a paired hardware keyboard), then restart the daemon and retry. Do not spend time on accessibility, Switch Control, Mirroring, or re-signing routes.
  • Setup uses Xcode. interceptor ios setup builds, signs, installs, and launches a team-scoped runner using the Apple Developer team configured in Xcode. interceptor ios login is unavailable and fails before password input. A background timer reruns Xcode setup before the certificate expires. Once registered, the runner stays resident for the XCUITest session; later verbs reuse it without a 30-second lease or relaunch.

Workflows

WorkflowWhen to invoke
workflows/drive-iphone-app.mdOpen an app and complete a task on a real iPhone: launch → tree → find → click/type → verify

References

FileTopic
references/command-catalog.mdFull interceptor ios command surface — setup, drive verbs, flags, addressing

When To Switch Surfaces

  • Target is content inside a browser tab (DOM, network, SPA state) → load interceptor-browser.
  • Target is a native macOS app (or the browser chrome / OS dialogs on the Mac) → load interceptor-macos.
  • Target is an app on the physical iPhone → this skill.

Do Not Default To Troubleshooting

  • User wants an iPhone task completed → run interceptor ios * commands.
  • User wants Interceptor's iOS support fixed, installed, or explained → that's a separate task; ask before diving into repo state.
  • Inside the Interceptor repo, use this skill for live device validation, not as the primary source of repo-development instructions.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Interceptor Ios AI skill do?

Drive any installed app on an owned, unlocked, Developer-Mode iPhone via interceptor ios *: ref-tagged element trees, deterministic coordinate taps (click), reliable text entry (type/keys), scroll, drag, multi-touch gestures (gesture), hardware buttons (press), screenshots, a continuous JPEG frame stream (stream/frame), installed-app listing, and app launch/activate/terminate. The phone runs an on-device XCUITest runner (InterceptorRunner) that dials into the daemon over WiFi — no cable once paired, no WebDriverAgent. Address a phone with --on <name> or ios:<udid>; phones auto-connect on th...

Why use Interceptor Ios on TypingMind?

Because you install it once and use it with any model. Interceptor Ios is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Interceptor Ios in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Hacker-Valley-Media/Interceptor/tree/main/.agents/skills/interceptor-ios. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Interceptor Ios?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Interceptor Ios?

As many as you like. As long as a model supports skills, you can use Interceptor Ios with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Interceptor Ios AI skill free?

It is published on GitHub by Hacker-Valley-Media. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇