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 NikiforovAll/claude-code-rules --skill nano-bananagit clone --depth 1 https://github.com/NikiforovAll/claude-code-rulesWrote 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/nikiforovall/claude-code-rules/nano-banana)<a href="https://agentmods.dev/skills/nikiforovall/claude-code-rules/nano-banana"><img src="https://agentmods.dev/badge/skills/nikiforovall/claude-code-rules/nano-banana/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/nikiforovall/claude-code-rules/nano-banana"><img src="https://agentmods.dev/badge/skills/nikiforovall/claude-code-rules/nano-banana.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00073 | $0.01304 |
| Opus 5 | $0.00036 | $0.00652 |
| Sonnet 5 | $0.00015 | $0.00261 |
| Haiku 4.5 | $0.00007 | $0.00130 |
Grade A, and why
nano-banana 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 10d 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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nano Banana Skill
Python scripting with Gemini image generation using uv. Write small, focused scripts using heredocs for quick tasks—no files needed for one-off operations.
Choosing Your Approach
Quick image generation: Use heredoc with inline Python for one-off image requests.
Complex workflows: When multiple steps are needed (generate -> refine -> save), break into separate scripts and iterate.
Scripting tasks: For non-image Python tasks, use the same heredoc pattern with uv run.
Writing Scripts
Execute Python inline using heredocs with inline script metadata for dependencies:
uv run - << 'EOF'
# /// script
# dependencies = ["google-genai", "pillow"]
# ///
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-3-pro-image-preview",
contents=["A cute banana character with sunglasses"],
config=types.GenerateContentConfig(
response_modalities=['IMAGE']
)
)
for part in response.parts:
if part.inline_data is not None:
image = part.as_image()
image.save("tmp/generated.png")
print("Image saved to tmp/generated.png")
EOF
The # /// script block declares dependencies inline using TOML syntax. This makes scripts self-contained and reproducible.
Why these dependencies:
google-genai- Gemini API clientpillow- Required for.as_image()method (converts base64 to PIL Image) and saving images
Only write to files when:
- The script needs to be reused multiple times
- The script is complex and requires iteration
- The user explicitly asks for a saved script
Basic Template
uv run - << 'EOF'
# /// script
# dependencies = ["google-genai", "pillow"]
# ///
from google import genai
from google.genai import types
client = genai.Client()
# Generate image
response = client.models.generate_content(
model="gemini-3-pro-image-preview",
contents=["YOUR PROMPT HERE"],
config=types.GenerateContentConfig(
response_modalities=['IMAGE']
)
)
# Save result
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("tmp/output.png")
print("Saved: tmp/output.png")
EOF
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 198 lines · 73 tokens per session scan A 9d83943344cf
nano-banana is a skill published in the GitHub repository NikiforovAll/claude-code-rules (141 stars, last pushed 11d ago), licensed Apache-2.0. It adds 73 tokens to every session and 1,304 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 2026-08-30.
Other skills, from other repositories
seedance-prompter
Expert prompt engineering for Seedance 2.0. Use when the user wants to generate a video with multimodal assets (images, videos, audio) and needs the best possible prompt.
cloudflare-workers-ai
Cloudflare Workers AI for serverless GPU inference. Use for LLMs, text/image generation, embeddings, or encountering AIERROR, rate limits, token exceeded errors.
structured-output
This skill should be used when the user mentions "structured output", "json mode", "tool schema", "function calling", "json parse error", "responseformat", "pydantic ai", "invalid json from the model", or "the model's JSON keeps breaking". It provides a methodology for getting reliably valid, schema-conformant…
prompt-versioning-hygiene
This skill should be used when the user mentions "prompt engineering", "prompt versioning", "prompt management", "system prompt", "prompt template", "manage prompts", or "prompts scattered in code". It provides a methodology for treating prompts as versioned, reviewable, testable code rather than magic strings buried…
mkprompt-goal
A goal-focused prompt builder that creates a short instruction around the desired outcome, how to check it, and when the work is objectively finished. It can also prepare prompts for work that continues across multiple sessions.
fal-prompt
Reçeteleri hedef modelin konuştuğu dile çevirme — model şemasından ve resmi örneklerinden prompt lehçesi çıkarma, negatif prompt ve parametre ayarı. fal-promptsmith agent'ı bunu okur. Use for writing prompts for a specific fal model, prompt format detection, negative prompts, seed and parameter tuning.