agy-skill-installer
Skill1 starsChanged 5 months agoMITView on GitHub
Install any agent skill into Anti-Gravity IDE with 100% fidelity, automated YAML frontmatter validation, dual-root provisioning, security scanning, and slash command integration.
In this repository19 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- apex-bug-hunterSkill
- arsSkill
- browser-triageSkill
- claude-md-initSkill
- delink-githubSkill
- gemini-skill-packagerSkill
- gh-readySkill
- indian-legal-verificationSkill
- local-systematic-verification-ocrSkill
- magento-team-leadSkill
- merge-gateSkill
- motioncraftSkill
- pdf-vector-engineSkill
- presscraftSkill
- score2mdSkill
- skill-shipSkill
- wpcraftSkill
The file
Files
Anti-Gravity Skill & Plugin Installer
Install any skill, multi-skill collection, or native plugin into Anti-Gravity IDE with 100% fidelity. Zero truncation, zero omitted subfolders, zero broken dependencies.
Overview
Anti-Gravity IDE supports two primary customization constructs:
- Agent Skills: Procedural runbooks and commands discovered across the 4 discovery roots (
~/.gemini/config/skills/,.agents/skills/,~/.gemini/antigravity/skills/, and~/.agents/skills/). - Native Antigravity Plugins (The Ponytail Standard): Namespaced bundles (
~/.gemini/config/plugins/<name>/and.agents/plugins/<name>/) packaging skills,AGENTS.mdrules, hooks, and MCP servers into an authoritative unit that is registered in~/.gemini/config/config.jsonwith direct toggle control ("enabled": true / false) inside the Antigravity Customization Settings UI.
Operating Protocol
A. Native Antigravity Plugin Installation (Ponytail Standard)
- Detection: Automatically identifies plugin repositories by inspecting
plugin.json,gemini-extension.json, or.antigravity-plugin/plugin.json, or via explicit-AsPluginflag. - Authoritative Deployment:
- Deploys full plugin package to Global (
~/.gemini/config/plugins/<name>/) and Local (.agents/plugins/<name>/). - Ensures valid
plugin.jsondeclaring plugin identity at the root. - Ensures
AGENTS.mdis preserved at the plugin root for runtime rules discovery.
- Deploys full plugin package to Global (
- Customization Settings Control Registration:
- Atomically registers
plugins.<name> = { "enabled": true }into~/.gemini/config/config.json. - Exposes native toggle switches in Antigravity IDE -> Settings -> Customizations -> Plugins, allowing users to enable or disable the entire plugin and all its capabilities instantly.
- Atomically registers
- Bundled Skill Encapsulation (Zero Duplicate & Token Optimization):
- Preserves bundled skills exclusively within
plugins/<name>/skills/. Antigravity's plugin engine automatically discovers and mounts them with the plugin badge without duplicating them into standalone skills roots, eliminating context token waste and ensuring clean lifecycle control when toggled on/off in Settings.
- Preserves bundled skills exclusively within
B. Standalone Skill Installation
- Source Parsing: Clones Git repository or resolves local directory.
- YAML Normalization: Fixes unquoted colons, invalid block scalars, and ensures
name:matches the directory name exactly. - Security & Privacy Scanning: Scans for leaked keys and tracking beacons.
- Multi-Root Provisioning: Copies files with 100% fidelity to:
- Global:
~/.gemini/config/skills/<name>/ - Local:
.agents/skills/<name>/ - Antigravity Root:
~/.gemini/antigravity/skills/<name>/ - User Agent Root:
~/.agents/skills/<name>/
- Global:
- Slash Command Registration: Automatically registers
/<name>in the chat interface.
Reusable Automation Scripts
- Windows (PowerShell) - Standalone Skill:
powershell -ExecutionPolicy Bypass -File "skills/agy-skill-installer/scripts/install-skill.ps1" -SourceUrlOrPath "<REPO_URL>" - Windows (PowerShell) - Native Antigravity Plugin (Ponytail Standard):
powershell -ExecutionPolicy Bypass -File "skills/agy-skill-installer/scripts/install-skill.ps1" -SourceUrlOrPath "<REPO_URL>" -AsPlugin - Linux / macOS (Bash):
bash skills/agy-skill-installer/scripts/install-skill.sh "<REPO_URL>" [--plugin]
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

