Skip to content

Docs/added installation instructions for Google Antigravity users to the README. #281

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
codeF1x wants to merge 3 commits into
base: main
Choose a base branch
from
+27 −8
Open

Docs/added installation instructions for Google Antigravity users to the README. #281

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c35ba51
docs: add Google Antigravity installation instructions
codeF1x Jan 17, 2026
bfd34a1
docs: add link to official Antigravity Skills documentation
codeF1x Jan 17, 2026
c8c70af
docs: use generic windows path variable
codeF1x Jan 17, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 27 additions & 8 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,26 +4,24 @@ Superpowers is a complete software development workflow for your coding agents,

## How it works

It starts from the moment you fire up your coding agent. As soon as it sees that you're building something, it *doesn't* just jump into trying to write code. Instead, it steps back and asks you what you're really trying to do.
It starts from the moment you fire up your coding agent. As soon as it sees that you're building something, it _doesn't_ just jump into trying to write code. Instead, it steps back and asks you what you're really trying to do.

Once it's teased a spec out of the conversation, it shows it to you in chunks short enough to actually read and digest.
Once it's teased a spec out of the conversation, it shows it to you in chunks short enough to actually read and digest.

After you've signed off on the design, your agent puts together an implementation plan that's clear enough for an enthusiastic junior engineer with poor taste, no judgement, no project context, and an aversion to testing to follow. It emphasizes true red/green TDD, YAGNI (You Aren't Gonna Need It), and DRY.
After you've signed off on the design, your agent puts together an implementation plan that's clear enough for an enthusiastic junior engineer with poor taste, no judgement, no project context, and an aversion to testing to follow. It emphasizes true red/green TDD, YAGNI (You Aren't Gonna Need It), and DRY.

Next up, once you say "go", it launches a *subagent-driven-development* process, having agents work through each engineering task, inspecting and reviewing their work, and continuing forward. It's not uncommon for Claude to be able to work autonomously for a couple hours at a time without deviating from the plan you put together.
Next up, once you say "go", it launches a _subagent-driven-development_ process, having agents work through each engineering task, inspecting and reviewing their work, and continuing forward. It's not uncommon for Claude to be able to work autonomously for a couple hours at a time without deviating from the plan you put together.

There's a bunch more to it, but that's the core of the system. And because the skills trigger automatically, you don't need to do anything special. Your coding agent just has Superpowers.


## Sponsorship

If Superpowers has helped you do stuff that makes money and you are so inclined, I'd greatly appreciate it if you'd consider [sponsoring my opensource work](https://github.com/sponsors/obra).

Thanks!
Thanks!

- Jesse


## Installation

**Note:** Installation differs by platform. Claude Code has a built-in plugin system. Codex and OpenCode require manual setup.
Expand Down Expand Up @@ -77,6 +75,23 @@ Fetch and follow instructions from https://raw.githubusercontent.com/obra/superp

**Detailed docs:** [docs/README.opencode.md](docs/README.opencode.md)

### Google Antigravity

For more details on managing skills, please refer to the [official documentation](https://antigravity.google/docs/skills).

1. Clone this repository:

```bash
git clone https://github.com/obra/superpowers temp_superpowers
```

2. Copy the `skills` directory contents to your global skills directory:

- **Windows**: `%USERPROFILE%\.gemini\antigravity\skills`
- **Mac/Linux**: `~/.gemini/antigravity/skills`

3. (Optional) Clean up the temporary directory.

Comment on lines 78 to 94
Copy link

@coderabbitai coderabbitai bot Jan 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

🧩 Analysis chain

🌐 Web query:

Google Antigravity skills installation documentation directory path

💡 Result:

Common Antigravity skill locations:

  • Workspace (project) scope: /.agent/skills/.[1][2]
  • Global (user) scope (official doc): ~/.gemini/antigravity/skills/.[1][2]
  • Note: some users report the runtime actually looks in ~/.gemini/antigravity/global_skills/ (create both if unsure). [3]

Sources: [1] Antigravity docs / Getting Started & Skills guides. [2] Medium tutorial on Antigravity Skills. [3] Community report showing the global_skills discrepancy.

🌐 Web query:

Google Antigravity .gemini directory configuration

💡 Result:

Short answer — Antigravity stores its local state and user/config artifacts in a .gemini folder (either in your project workspace and/or your home directory), and you can control agent behavior with simple files there (notably GEMINI.md plus subfolders like antigravity/brain and artifacts). See sources below for examples and community notes.

Quick facts / where to look

  • Locations reported by users: ~/.gemini/antigravity (home) and a .gemini folder inside project workspaces. [1][2]
  • What’s inside: artifacts, conversation/chat data, “brain” memory, and configuration files used by Antigravity agents. Users report task lists, artifacts and chats are stored under .gemini/antigravity/brain and related subfolders. [1][3]
  • GEMINI.md: a plaintext Markdown file in the .gemini directory that many users place rules/instructions into (eg. agent safety rules, “never auto-execute” style constraints). Antigravity will read that file as a source of agent instructions/constraints. [1][4]
  • Related config examples: some community tools and opencode plugins show Antigravity-related model/provider config and naming conventions (e.g., antigravity- prefixes and model entries) in JSON config (opencode example). That config is separate from the .gemini runtime folder but helpful when integrating CLI/plugins. [5]

Practical tips

  • Check both ~/.gemini and project-root/.gemini if you can’t find artifacts. User reports show both places are used. [1][3]
  • Back up .gemini before experiments — users report lost/vanishing conversations when reinstalling or switching models. [3]
  • To add global agent rules, create/edit .gemini/GEMINI.md with line-by-line constraints (e.g., “NEVER execute terminal commands without explicit confirmation”). Keep it plain text/Markdown. [1]

Sources

  1. User reports and GEMINI.md examples — Reddit threads discussing .gemini, GEMINI.md, and artifact locations. [reddit.com threads].
  2. Reports of Antigravity storing chats/artifacts in ~/.gemini/antigravity. [reddit.com].
  3. Thread reporting artifacts & “brain” folder location in .gemini. [reddit.com].
  4. News/overview of Google Antigravity (agent-first IDE) — context on features that write artifacts. (The Verge).
  5. opencode GitHub example showing Antigravity-related model/provider JSON configuration patterns.

If you want, tell me your OS and whether you’re looking for the project-local .gemini or the user ~/.gemini and I’ll show exact paths and an example GEMINI.md template you can drop in.

🌐 Web query:

site:google.com Antigravity skills setup

💡 Result:

You can set up Antigravity (Google’s Antigravity IDE/agents) and configure skills by following Google’s getting-started codelab and Cloud blog guidance. Key steps:

  • Install Antigravity and sign in with a Google account, then run the Setup Wizard (includes editor preferences, browser extension, and optional CLI). (codelabs.developers.google.com)
  • Configure agent “skills” via the Agent Manager / Editor: choose models, enable browser subagent and extensions, and set editor keybindings and extensions. (codelabs.developers.google.com)
  • Connect data and skill integrations using MCP servers (install MCP connectors for BigQuery, AlloyDB, Cloud SQL, etc.) so agents can access your data as skills. (cloud.google.com)
  • Use the security controls (Terminal Auto-Execution policy, Allow/Deny lists, and browser permissions) to safely permit agent actions. (codelabs.developers.google.com)

If you want, I can open the specific Google codelab or Cloud blog page and extract the exact commands and UI paths for your OS and account—shall I do that?

Citations:

Fix the documentation URL for Google Antigravity.

The link https://antigravity.google/docs/skills does not appear to be a valid Google Antigravity documentation URL. According to official Google sources, use the codelab at https://codelabs.developers.google.com/getting-started-google-antigravity instead.

Additionally, the Windows path hardcodes C:\Users\Administrator\.gemini\antigravity\skills, which assumes a specific username. Use an environment variable like %USERPROFILE% to make it user-agnostic (e.g., %USERPROFILE%\.gemini\antigravity\skills).

🤖 Prompt for AI Agents 
In `@README.md` around lines 78 - 94, Update the "Google Antigravity" README
section by replacing the invalid docs link with the official codelab URL (use
https://codelabs.developers.google.com/getting-started-google-antigravity) and
make the Windows skills path user-agnostic by changing the hardcoded
C:\Users\Administrator\.gemini\antigravity\skills to use the %USERPROFILE%
environment variable (e.g., %USERPROFILE%\.gemini\antigravity\skills) so the
"Google Antigravity" section and its bullet points are corrected.

## The Basic Workflow

1. **brainstorming** - Activates before writing code. Refines rough ideas through questions, explores alternatives, presents design in sections for validation. Saves design document.
Expand All @@ -100,13 +115,16 @@ Fetch and follow instructions from https://raw.githubusercontent.com/obra/superp
### Skills Library

**Testing**

- **test-driven-development** - RED-GREEN-REFACTOR cycle (includes testing anti-patterns reference)

**Debugging**

- **systematic-debugging** - 4-phase root cause process (includes root-cause-tracing, defense-in-depth, condition-based-waiting techniques)
- **verification-before-completion** - Ensure it's actually fixed

**Collaboration**
**Collaboration**

- **brainstorming** - Socratic design refinement
- **writing-plans** - Detailed implementation plans
- **executing-plans** - Batch execution with checkpoints
Expand All @@ -118,6 +136,7 @@ Fetch and follow instructions from https://raw.githubusercontent.com/obra/superp
- **subagent-driven-development** - Fast iteration with two-stage review (spec compliance, then code quality)

**Meta**

- **writing-skills** - Create new skills following best practices (includes testing methodology)
- **using-superpowers** - Introduction to the skills system

Expand Down