MCP servers and skills: search, filter by where they run, and compare.
⚡ Token-Efficient multi-agent orchestration for Google Antigravity. Pairs Gemini 4 Argon (architect) with parallel Gemini 3.8 Flash subagents for 65%+ token savings. Implements DeepMind Pattern #4.
SkillMIT★ 4Few starsNo key
Antigravity plugin that pins Gemini 4 Argon (High or Medium) as the primary orchestrator and dynamically pins subagents to flash or flash_lite based on context
SkillMIT★ 2Few starsNo key
Use Gemini as an agent in Codex through Antigravity CLI, with selectable models and Google AI subscription login, without API keys or pay-as-you-go API billing
SkillGPL-3.0★ 3Few starsNo key
Inputs, prompts, scripts and outputs behind RanThis hands-on tests of AI tools and Claude skills.
SkillMITFew starsNo key
Multi-agent skill for Claude Code, Codex, and Agent Skills hosts: run t() and p() teams of AI coding agents in tmux. Write the team. Your agents take the turns.
SkillApache-2.0★ 1Few starsNo key
MCP bridge that drives the official Antigravity CLI (agy) — use your Antigravity subscription models from WorkBuddy, Claude Code, Cursor or any MCP client.
SkillMIT★ 1Few starsNo key
Claude Code plugin that delegates coding and review tasks to Vertex AI Gemini via the Antigravity SDK and google-genai, with isolated worktrees, HMAC receipts and provenance hooks.
SkillApache-2.0★ 1Few starsNo key
A benchmark and open call for valuable, reproducible RL runs.
SkillApache-2.0★ 13No key
Fifteen Markdown skills for coding agents. Copy a folder, or clone the repo and point skill discovery at skills/.
SkillMIT★ 1Few starsNo key
Cross-model frontier orchestrator for Google Antigravity (agy), Claude Code and OpenAI Codex CLI: a frontier primary plans and integrates while haiku/flash-lite scouts and sonnet/flash workers do look
SkillMIT★ 4Few starsNo key
Multi-model LLM council (karpathy/llm-council style) over the agent CLIs you already pay for: Codex, Antigravity, OpenCode, Claude Code. Stdlib Python, Windows-ready, Claude Code skill.
SkillMIT★ 3Few starsNo key
Universal JIT Skills & Hierarchical Context Tiering Optimizer (L1/L2/L3) for AI Coding Agents
SkillMIT★ 2Few starsNo key
STA (Supremo Tribunal dos Agentes). Corte Suprema adversarial no terminal: julgamento dialético em 2 rodadas (Defesa vs Acusação -> Magistrado) para código, negócios e decisões críticas.
SkillMITFew starsNo key
Practical, visual, example-driven guide to Google Antigravity (App, CLI, Python SDK) & Gemini (English + Tiếng Việt)
SkillMITFew starsNo key
Token Saver: Automating the Model Picker. Open-source Agent Skills that send each AI task to the cheapest model that can finish it, for Anthropic, OpenAI, Google, DeepSeek, Qwen and Perplexity. 13 per
SkillMITFew starsNo key
Who stands behind it
