Omahub
← All plugins
Y

Antigravity Voice Dictation

by yusuf

Seamless global voice-to-text dictation using Antigravity AI in Hyprland.

GitHub

Security review

Review recommended · 1 finding

Deterministic scan — not a security guarantee

Low
Risk level
Low
Analyzed commit
db26c41
Scanned
1 month ago

Flagged patterns appear only in documentation files (README / docs) — descriptive examples, not executable code.

  • Docs external_hosts README.md:46

    Downloads or connects to an external HTTP(S) host.

    git clone https://github.com/yusufalvian16/antigravity-dictation.git

Automated analysis only — not a security guarantee.

AI advisory review

No obvious issues detected

Language-model assessment · ~deepseek/deepseek-v4-flash-latest — advisory only

Low
AI risk level
Low
Recommendation
install
Model
~deepseek/deepseek-v4-flash-latest
Analyzed commit
db26c41
Reviewed
1 month ago

This is a legitimate voice dictation automation script that toggles Antigravity IDE's dictation mode, copies the transcribed text, and restores window focus. The code is well-written with defensive security practices (address validation, atomic file writes, bounded buffers), and the only flagged external host is a git clone URL in the README documentation. The Service.qml entry point is essentially empty, with all functionality living in the clearly-documented shell script.

  • The install.sh modifies ~/.config/hypr/bindings.lua to add a Super+D keybinding, which is documented behavior but does modify user config without explicit confirmation
  • The script uses non-standard 'hl.dsp.*' Hyprland dispatchers which may be custom or environment-specific, but this is not malicious
  • The script simulates keystrokes via wtype (Ctrl+M, Ctrl+A, Ctrl+C), which is expected for a dictation tool but could theoretically be misused if the script were ever tampered with
How this check works

This review combines the deterministic scan (the rule-based results above) with an independent look at the plugin's code by a language model. The model reads a trimmed sample of the repository's files, the manifest, and the README, then gives a plain-language risk level and a recommendation: install (no notable danger), review (look closer first), or avoid (clearly dangerous).

It runs on the same analyzed commit as the deterministic scan and is strictly advisory — it is not a security guarantee and never blocks a plugin by itself. A human moderator still reviews plugins before they are listed.

AI advisory only — automated analysis, not a security guarantee.

Install
$ omarchy plugin add https://github.com/yusufalvian16/antigravity-dictation --enable
Productivity #Hyprland #bar #ai

Antigravity Dictation

Omarchy Shell Plugin Hyprland License: MIT

A lightweight global voice-to-text dictation workflow for Omarchy Linux and Hyprland, powered by Antigravity AI.

Instead of running heavy local models that consume high CPU resources, this plugin routes speech recognition through Antigravity's fast, cloud-assisted backend.

Features

  • Zero Local CPU Overhead: Speech recognition runs via Antigravity without straining local hardware.
  • Multilingual Support: High accuracy for Indonesian, English, mixed phrasing, and automatic punctuation.
  • Workspace-Agnostic: Operates seamlessly whether Antigravity is in a scratchpad or across regular workspaces.
  • Native OSD Feedback: Uses Omarchy's transient On-Screen Display without polluting notification history.
  • Automatic Focus Restoration: Returns focus to your previous active window upon completing dictation.
  • Keyboard-Driven: Operates via a single global toggle shortcut (Super + D).

Requirements

Ensure the following prerequisites are met before use:

  • Omarchy Linux with Hyprland and Quickshell (omarchy-shell).
  • Antigravity IDE installed, running, and signed in to an active account.
  • Voice input working via Ctrl + M inside the Antigravity chat input.
  • System dependencies: wtype, jq, and hyprctl (pre-installed on Omarchy).

Installation

Method 1: Using Omarchy CLI (Recommended)

omarchy plugin add https://github.com/yusufalvian16/antigravity-dictation

Method 2: Manual Installation

git clone https://github.com/yusufalvian16/antigravity-dictation.git
cd antigravity-dictation
./install.sh

The installer copies the script to ~/.local/bin/antigravity-dictation, registers the plugin, and configures the Super + D keybinding in ~/.config/hypr/bindings.lua.

Usage

The dictation workflow is controlled using the Super + D shortcut:

sequenceDiagram
    autonumber
    actor User
    participant App as Active Window
    participant Core as Antigravity Dictation
    participant AGY as Antigravity

    User->>Core: Press Super + D
    Core->>App: Save current window address
    Core->>AGY: Focus Antigravity & activate mic (Ctrl + M)
    Note over User,AGY: User speaks. Transcription renders in Antigravity.

    User->>Core: Press Super + D again
    Core->>AGY: Stop mic (Ctrl + M) & copy text (Ctrl + A, Ctrl + C)
    Core->>App: Restore focus to original active window
    Core->>User: Display Copied OSD (Ready to paste with Ctrl + V)
  1. Press Super + D while focused on any application. The OSD displays Recording....
  2. Speak your text.
  3. Press Super + D again when finished. The OSD displays Processing... followed by Copied!.
  4. Paste the text into your target application using Ctrl + V.

Configuration

To change the default shortcut, edit ~/.config/hypr/bindings.lua:

o.bind("SUPER + D", "Antigravity Voice Dictation", "antigravity-dictation")

Uninstallation

To remove the plugin, binary, and keybindings cleanly:

cd antigravity-dictation
./uninstall.sh

License

Distributed under the MIT License.