Interceptor logo

Interceptor

Organization
Hacker-Valley-Media
interceptor

Choose the right Interceptor surface. Use interceptor-browser for page DOM, network, browser tabs, rich editors, screenshots, and browser automation. Use interceptor-macos for native apps, browser chrome, URL bars, OS dialogs, cross-app routing, AX trees, native screenshots, Apple Events, trusted OS input, macOS Electron CDP/app web-content control, and in-process app runtime control. Use interceptor-ios for apps on an owned, unlocked, Developer-Mode iPhone. Use interceptor-research for deep multi-source web research. Background-first by default; focus changes require explicit opt-in.

Overview

PublisherHacker-Valley-Media
RepositoryInterceptor
Skill nameinterceptor
Stars
526
Forks
76
Bundled files
Instructions only
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

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

Installation

Install the Interceptor 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
  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 .claude/skills/interceptor
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Interceptor 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 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 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

Use this as the routing skill before loading a surface-specific skill.

Surface Decision

TaskSkill
Page DOM, text, network, SPA state, browser monitor, screenshots of browser contentinterceptor-browser
Rich browser editors (Docs/Slides/Canva), canvas/WebGL camera viewers, scene-graph reads, page-world request overrides, blob/export capture, or "cooking" a live page with overlays/HUDsinterceptor-browser
Native macOS apps, OS dialogs, browser chrome, URL bars, app windows, menu barsinterceptor-macos
Electron / Chromium desktop app web contents, such as Slack, VS Code, Notion, or Descript DOM/network/JSinterceptor macos cdp / interceptor macos cdp app via interceptor-macos/references/cdp-app.md
Native app runtime internals: live view/object graph, selector calls, rendered text mutation, hooks, MapKit/AppKit/SwiftUI runtime workinterceptor macos runtime via interceptor-macos/references/native-agent.md
Open or control a named app such as Brave, Mail, Finder, Signal, or Cursorinterceptor-macos
Backgrounded, occluded, minimized, or cross-Space app captureinterceptor-macos
Deep web research: investigate a topic across many sources with breadth + verificationinterceptor-research
Owned physical iPhone, app automation, runner state and device servicesinterceptor-ios
Capabilities added by an installed extension (operator-supplied)run interceptor extensions list, then load the extension's own skill (interceptor-ext-<name>)

The User Is On This Mac Too

Assume the person is working on this Mac while you work. Stay in the background at all costs. Take focus (the frontmost app, the focused window, or the tab showing in a window the user is using) only when the action truly cannot be done any other way. Never take it to make your own work easier, and never to test something.

Before any focus-moving verb, go down this list and stop at the first rung that works:

  1. A background verb. Synthetic browser input, AX refs, --app / --pid / --window on interceptor macos input, and --trusted browser input (a full install delivers it to an unfocused window).
  2. interceptor tab keepalive <id> for a page that will not render while hidden.
  3. Your own window. interceptor window new [url] moves your tab group into its own background window. Tabs there paint, play media, and take trusted clicks, and interceptor tab switch <id> there changes only that window. Never switch tabs in a window the user is working in.
  4. Only then a focus move. Say so first, keep it to one fast batch, and put back the frontmost app, the active tab, and the cursor.

Known foreground-only cases: launching an app that is not running (it can come forward even from a background launch, and an app updater can raise a password prompt), and a click or drag on a minimized window or hidden app.

Core Rules

  • Browser commands operate inside the cyan interceptor tab group. Do not use --any-tab unless the user explicitly authorizes acting outside that group.
  • Solo browser work in a supported agent shell gets a soft per-session group automatically. For concurrent lanes, set a unique INTERCEPTOR_SESSION_ID or pass a unique --group <label> to each lane. Explicit groups are hard-scoped by default unless --any-tab is authorized. Close named groups with interceptor group close <label> when done; the idle timer is based on tab activity, so metadata polls do not keep a group alive.
  • interceptor open <url>, interceptor tab new <url>, and interceptor window new [url] create background tabs and windows by default. Only open --activate, tab new --activate, window new --activate, tab switch <id>, and window focus <id> intentionally move browser focus.
  • A page that will not render while hidden is not a reason to switch tabs in the user's window. interceptor tab keepalive <id> makes it read as visible and run its animation frames in place. A page that needs a real paint (video playback, a virtualized grid that measures itself) goes in your own window: interceptor window new <url>.
  • A trusted click does not need a tab switch or a focused window: interceptor click <ref> --trusted reaches the active tab of any window that is not minimized. If your tab is not the active tab of its window, interceptor window new then tab switch <id> inside that window.
  • The macOS surface is background-first by default. Only interceptor macos app activate <app> and interceptor macos open <app> --activate intentionally move focus.
  • If multiple browser profiles are connected, run interceptor contexts and pass --context <id>.
  • Safari registers as safari; use interceptor --context safari <verb> for page content and interceptor macos for Safari chrome or native fallbacks.
  • Prefer compound commands (open, read, act, inspect) and structured reads before screenshots.
  • The zero-CDP browser rule governs the user's real Chrome/Brave/Safari web session. For owned Electron apps, interceptor macos cdp and interceptor macos cdp app are intentional app-control surfaces.
  • For native app runtime internals, use interceptor macos runtime after checking interceptor status; public Full installs may require operator-supplied runtime agent dylibs/signing identity before macos runtime enable.
  • If the extension behaves stale after a package update, run interceptor reload --context <id>: an unpacked copy picks up the installed files; a Chrome Web Store copy only asks the store for an update and keeps its version until the store publishes the new one. interceptor contexts --verbose shows which copy (store or unpacked) and version each context runs.
  • interceptor daemon stop and interceptor skills unadopt are install/maintenance verbs. The daemon is shared by every agent on the machine — never stop it during normal automation; the next browser command respawns it, but every connected agent's in-flight work is dropped.

Load A Surface Skill

  • Load interceptor-browser for browser page content, network, tabs, scene graphs, and browser screenshots.
  • Load interceptor-macos for native apps, browser chrome, OS dialogs, window capture, AX trees, Apple Events, Electron app CDP/app attach, and in-process native agent control.
  • Load interceptor-ios for an owned, unlocked, Developer-Mode iPhone. Disconnected unlock cannot bootstrap its runner.
  • Load interceptor-research for deep web research — investigating a topic across many sources with a planner loop, an on-disk source ledger, and adversarial verification (methodology layered on the browser surface). Pull the playbook any time with interceptor research.

Frequently asked questions

What does the Interceptor AI skill do?

Choose the right Interceptor surface. Use interceptor-browser for page DOM, network, browser tabs, rich editors, screenshots, and browser automation. Use interceptor-macos for native apps, browser chrome, URL bars, OS dialogs, cross-app routing, AX trees, native screenshots, Apple Events, trusted OS input, macOS Electron CDP/app web-content control, and in-process app runtime control. Use interceptor-ios for apps on an owned, unlocked, Developer-Mode iPhone. Use interceptor-research for deep multi-source web research. Background-first by default; focus changes require explicit opt-in.

Why use Interceptor on TypingMind?

Because you install it once and use it with any model. Interceptor 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 in TypingMind?

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

Which AI models can use Interceptor?

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?

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

Is the Interceptor 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 👇