Ultimate Gemini Skill banner
bradAGI bradAGI

Ultimate Gemini Skill

Development community

Description

Standalone Claude Code skill for Gemini 3.1 Flash Image — CLI mirror of ultimate-image-gen-mcp, no MCP server required

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

README

ultimate-gemini-skill

![banner](banner.png)

🚀 Easiest way to install

One command — no git clone, no manual copying:

npx skills add bradAGI/ultimate-gemini-skill

That's it. Auto-detects your coding agent and drops the skill in the right place.

Works with every agent supported by [skills.sh](https://skills.sh)

This skill is agent-agnostic. The `npx skills add` installer supports **19+ coding agents / IDEs**, including:

**Claude Code** · **Cursor** · **Windsurf** · **Codex** · **Cline** · **GitHub Copilot** · **Gemini** · **VSCode** · **Goose** · **AMP** · **Antigravity** · **ClawdBot** · **Droid** · **Kilo** · **Kiro CLI** · **Nous Research** · **OpenCode** · **Roo** · **Trae**

If your agent loads skills from a standard `skills/` or similar directory, the installer will place the files correctly. You can also install manually (see below) — the skill itself is just a `SKILL.md` + a Python CLI, so any harness that can read a markdown skill definition can use it.

After installing, set `GEMINI_API_KEY` (see [Securing your API key](#securing-your-api-key) below) and run `pip install google-genai pillow` once for the bundled CLI.


A **standalone, agent-agnostic** skill for Gemini 3.1 Flash Image (Nano Banana 2). Ships a self-contained Python CLI — `scripts/gemini_image.py` — that calls the Gemini API directly. Derived from [`ultimate-image-gen-mcp`](https://github.com/anand-92/ultimate-image-gen-mcp) by [@anand-92](https://github.com/anand-92) — all credit for the underlying tool design and prompt templates goes to the original author.

The banner above was generated by the skill itself — `generate-image --aspect-ratio 21:9 --image-size 1K`.

Samples

All three generated with this skill at **1K**, rendered by `scripts/gemini_image.py`. Commands shown.

1. Real-world grounded photo (Google Search + Image Search)

![tony hawk](samples/tony-hawk-grounded.png)

python3 scripts/gemini_image.py generate-image \
  --prompt "Ton