Skills nano-banana-2
🎨

nano-banana-2

Low Risk ⚙️ External commands🌐 Network access

Generate Images with Google Nano Banana 2

Also available from: inference-sh-9,tul-sh,inferen-sh,agentspace-so,runcomfy-com,inference-shell,toolshell

Quickly generate images with Google Nano Banana 2 via RunComfy CLI. Includes optimized prompting patterns, batch generation support, and web-grounded context for current events.

Supports: Claude Codex Code(CC)
⚠️ 64 Poor
1

Download the skill ZIP

2

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

3

Toggle on and start using

Test it

Using "nano-banana-2". A minimalist 16:9 product card: a matte black ceramic mug centered on a soft warm-grey paper background, rim highlight from upper-left, the headline "Brewed Quietly" in clean bold sans-serif top-right

Expected outcome:

A 16:9 PNG image showing a centered black ceramic mug on grey paper with "Brewed Quietly" text rendered in bold sans-serif at top-right, clean studio lighting, balanced composition

Using "nano-banana-2". A 9:16 vertical hero for a wellness brand: a single ceramic teacup on a linen runner, soft morning side-light, the words "Slow Down" in hand-drawn serif large at the top

Expected outcome:

A 9:16 vertical PNG showing a teacup on linen with "Slow Down" rendered in hand-drawn serif at top, soft natural lighting, warm neutral palette

Security Audit

Low Risk
v1 • 5/31/2026

Evaluated 88 static pattern detections. All high-severity findings are false positives: backtick patterns are markdown code formatting, path traversal is a markdown link reference, hidden file detection is a standard CLI config path, and weak crypto/system reconnaissance are keyword misidentifications. The skill is legitimate image generation documentation calling the RunComfy CLI. User prompts are transmitted as JSON without shell expansion, preventing injection. No malicious patterns confirmed.

1
Files scanned
207
Lines analyzed
8
findings
1
Total audits

High Risk Issues (3)

Path Traversal Sequence Flagged
Static analyzer flagged '../nano-banana-edit' at line 70 as path traversal. This is a markdown relative link reference to a sibling skill file, not a security vulnerability. FALSE POSITIVE.
Hidden File Access Flagged
Static analyzer flagged '~/.config/runcomfy/token.json' at line 202 as hidden file access. This is the documented RunComfy CLI token storage location with mode 0600 (owner-only). Legitimate configuration behavior, not a vulnerability. FALSE POSITIVE.
Weak Cryptographic Algorithm Flagged
Static analyzer flagged lines 4, 60, 132, 183 as weak crypto. Direct review shows no cryptographic algorithms are used - the analyzer misidentified documentation keywords. FALSE POSITIVE.
Medium Risk Issues (1)
Ruby/Shell Backtick Execution Pattern Detected
Static analyzer flagged 70 instances of backtick patterns. These are markdown code formatting syntax (`code` and ```bash blocks) in documentation, not Ruby/shell backtick execution. The code examples show legitimate CLI invocation examples. FALSE POSITIVE.
Low Risk Issues (2)
Hardcoded URLs Detected
Static analyzer flagged URLs to runcomfy.com and model-api.runcomfy.net. These are expected external service endpoints for this image generation skill. LEGITIMATE - no data exfiltration or suspicious endpoints.
System Reconnaissance Pattern Flagged
Static analyzer flagged mentions of 'Gemini' model family as system reconnaissance. These are legitimate product references in documentation describing the Nano Banana 2 model lineage. FALSE POSITIVE.

Risk Factors

⚙️ External commands (1)
🌐 Network access (1)
Audited by: claude

Quality Score

38
Architecture
100
Maintainability
87
Content
50
Community
51
Security
83
Spec Compliance

What You Can Build

Rapid Marketing Draft Iterations

Generate batch of 4 thumbnail variants at 0.5K resolution for quick visual exploration, then promote winning variant to 2K for final delivery.

Social Media Asset Creation

Create platform-native assets with 9:16 vertical format for Reels/Stories, 4:5 for Instagram feed, or 1:1 for posts - with predictable in-image text rendering.

Current Event Visualizations

Use web-grounded generation with enable_web_search to create images referencing real-world entities or this week's news events.

Try These Prompts

Basic Image Generation
Use nano-banana-2 to generate: A minimalist product photo of a white ceramic mug on a grey background with the text "Hello World" in bold sans-serif, centered
Batch Variant Generation
Use nano-banana-2 to generate 4 vertical variants of: A wellness brand hero showing a ceramic teacup on linen, with the words "Slow Down" in hand-drawn serif, soft morning light
Web-Grounded Current Event Image
Use nano-banana-2 with web search enabled to generate: A cinematic shot of the SpaceX launch from yesterday, rocket against orange sunset sky
Reproducible Seed-Locked Variation
Use nano-banana-2 with seed 42 to generate: A 16:9 cinematic portrait of a chef in a Paris kitchen, golden hour light through windows, ultra-detailed realistic

Best Practices

  • Front-load the primary subject, then add environment, style, and camera directives
  • Quote literal text characters exactly when you need in-image typography rendered
  • Start with 0.5K resolution for rapid iteration, upgrade to 2K only for final deliverables

Avoid

  • Conflicting style directives - do not combine "minimalist + ornate + retro + cyberpunk"
  • Non-quoted text for in-image typography - expect unpredictable rendering
  • Requesting resolutions outside 0.5K, 1K, 2K, 4K tiers - will return 422 error

Frequently Asked Questions

What triggers the nano-banana-2 skill?
Use phrases like "nano banana", "nano-banana-2", "nano banana 2", "google image gen", "gemini image", or any explicit request to generate with this model.
How do I render text inside an image?
Quote the exact text characters you want rendered. For example: "the label reads 'AURA' in clean bold sans-serif, centered".
Why are my generated images blurry at 0.5K?
0.5K is for rapid drafts and ideation only. Use 1K for standard output, or 2K/4K for final deliverables that need sharp detail.
When should I enable web search?
Only enable enable_web_search when your prompt references current events or real-world entities that need up-to-date information. It adds cost and latency.
Can I edit existing images with this skill?
No, this skill is for text-to-image generation only. For image editing (preserving subject, swapping background), use the separate nano-banana-edit skill.
How do I get consistent results for refinement?
Set the seed parameter to a fixed number (like 42) when iterating on a prompt. Reusing the same seed with the same prompt produces consistent composition.

Developer Details

File structure

📄 SKILL.md