Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add u2n4/gemini-media-mcp --skill nanobanana-promptinggit clone --depth 1 https://github.com/u2n4/gemini-media-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/u2n4/gemini-media-mcp/nanobanana-prompting)<a href="https://agentmods.dev/skills/u2n4/gemini-media-mcp/nanobanana-prompting"><img src="https://agentmods.dev/badge/skills/u2n4/gemini-media-mcp/nanobanana-prompting/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/u2n4/gemini-media-mcp/nanobanana-prompting"><img src="https://agentmods.dev/badge/skills/u2n4/gemini-media-mcp/nanobanana-prompting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00085 | $0.06341 |
| Opus 5 | $0.00043 | $0.03171 |
| Sonnet 5 | $0.00017 | $0.01268 |
| Haiku 4.5 | $0.00009 | $0.00634 |
Grade A, and why
nanobanana-prompting scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 604 lines — stays where its author put it; the contents beside it link to each section on GitHub.
NanoBanana Pro Prompting Mastery
When to Use This Skill
Activate this skill EVERY TIME before calling nanobanana:generate_image. Read and apply these techniques to transform any image request into a professional-grade prompt.
Prompt Format Selection (DECIDE FIRST)
Before writing any prompt, choose the right format based on scene complexity. There are 4 proven formats — each excels at different tasks:
Format Decision Matrix
| Complexity | Format | Best For | Example |
|---|---|---|---|
| ⭐ Simple | Plain Text | Single subject, one style, clear idea | Maps, icons, simple portraits, logos |
| ⭐⭐ Medium | Descriptive Narrative | Creative/artistic scenes with mood and atmosphere | Surreal art, vintage documents, fantasy scenes |
| ⭐⭐⭐ Complex | Structured Markdown | Technical realism with precise camera/lighting specs | Mirror selfies, phone-style photos, architectural |
| ⭐⭐⭐⭐ Advanced | JSON | Multiple subjects, fashion editorial, reusable templates | Multi-person scenes, product + model, infographics |
Golden Rule: More complex scene → Stronger structure needed.
Format 1: Plain Text (Simple & Fast)
Use for single-subject images with one clear style. No structure needed.
Generate a map of Germany in watercolor style, on which all
federal states are labeled in ballpoint pen.
Why it works: One subject + one style + one detail = the model gets it instantly with no confusion. Adding structure here would be overkill.
Use when: Maps, simple portraits, icons, single objects, quick concepts.
Format 2: Descriptive Narrative (Sensory Storytelling)
Use for creative/artistic scenes that need mood and atmosphere. Write like a film director describing a shot.
A vintage patent document for INVENTION, styled after late 1800s
United States Patent Office filings. The page features precise
technical drawings with numbered callouts (Fig. 1, Fig. 2, Fig. 3)
showing front, side, and exploded views. Handwritten annotations
in fountain-pen ink describe mechanisms. The paper is aged ivory
with foxing stains and soft fold creases. An official embossed seal
and red wax stamp appear in the corner. The entire image feels
like a recovered archival document—authoritative, historic, and
slightly mysterious.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 604 lines · 85 tokens per session scan A 92a75658a1b7
nanobanana-prompting is a skill published in the GitHub repository u2n4/gemini-media-mcp (1 stars, last pushed 2mo ago), licensed MIT. It adds 85 tokens to every session and 6,341 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
llm-app-patterns
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
prompt-optimization
Improve a prompt on the evaluations workbench through a measured loop. Score the baseline first, then duplicate the target column, form a hypothesis from failing rows, edit the copy's prompt draft, run, compare pass rate and cost, and repeat until the numbers hold. Use when the user asks to optimize or improve a…
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
prompt-engineer
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot…
seedance-vocab-en
This skill should be used when an English Seedance 2.0 prompt needs clearer production wording, less generic prose, or precise vocabulary for camera, lighting, motion, VFX, audio, and constraints. Route blocked prompts through seedance-filter for context and boundary review.
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…