rafaelghif/antigravity-agents

Autonomous L9 multi-agent engineering framework for Google Antigravity. Features epistemic grounding, strict AST verification gates, GraphRAG blast radius, 11 enterprise skills, and universal language support.

5Stars on the repository
22Mods indexed here, across every type
todayLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

architecture

01

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill for system design, distributed architecture, strict RFC 7807 API contracts, and resilience engineering (idempotency, outbox, circuit breaker).

not rated 5 changed today A 34 tokens original MIT

caveman

02

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill to minimize token consumption by 60%+ using terse, high-density telegraphic responses ("Mouth smaller, not brain smaller") while keeping technical code and commands 100% byte-exact.

not rated 5 changed today C 46 tokens original MIT

code-quality

03

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill when writing, refactoring, or reviewing code to ensure SOLID principles, clean architecture, DRY deduplication, and anti-overengineering simplification.

not rated 5 changed today A 37 tokens original MIT

data-engineering

04

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill for database schemas, zero-downtime expand-contract migrations, concurrent indexing, ETL/ELT data pipelines, and streaming CDC.

not rated 5 changed today A 34 tokens original MIT

design

05

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill for UI/UX components, CSS/Tailwind styling, WCAG 2.2 AA accessibility, design tokens, and Core Web Vitals performance optimization.

not rated 5 changed today A 37 tokens original MIT

devops

06

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill for Docker containers, Kubernetes manifests, CI/CD pipelines, Infrastructure as Code (Terraform), and Model Context Protocol (MCP) server setup.

not rated 5 changed today A 34 tokens original MIT

observability

07

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill when the user asks to implement logging, monitoring, metrics, or distributed tracing in the application.

not rated 5 changed today A 25 tokens original MIT

security

08

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill when the user asks to modify authentication, authorization, CI/CD pipelines, Dockerfiles, handling user input, or touching secrets.

not rated 5 changed today B 30 tokens original MIT

semantic-graphing

09

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill when you need to understand the architecture, classes, and function signatures of a large repository without reading files blindly.

not rated 5 changed today A 30 tokens original MIT

verification

10

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Use this skill when the user asks to run tests, or immediately after modifying code to execute rigorous test-driven validation.

not rated 5 changed today A 25 tokens original MIT

deep-research

11

rafaelghif/antigravity-agents

Skill Claude CodeCodex

Epistemic web research, documentation lookup, and external API verification to eliminate hallucinations.

not rated 5 today A 22 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: