nano-banana-2
Generate Images with Nano Banana 2
Image teams need fast drafts, platform formats, and reliable text rendering. This skill guides Nano Banana 2 prompts and RunComfy CLI settings.
Stop for confirmation before installing.
Review the plan and obtain explicit user consent before changing files.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "nano-banana-2" from https://skillstore.io/skills/doany-ai-nano-banana-2.md and its manifest at https://skillstore.io/api/skills/doany-ai-nano-banana-2/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "nano-banana-2". Create four vertical launch-image options for a wellness tea campaign.
Expected outcome:
Recommended setup: four images, 9:16 ratio, 0.5K resolution for quick review. Prompt focus: one ceramic teacup, morning light, quoted headline text, and uncluttered composition.
Using "nano-banana-2". Make a final 16:9 product card with the headline Brewed Quietly.
Expected outcome:
Recommended setup: one image, 16:9 ratio, 2K resolution, and a fixed seed. Prompt focus: centered product, precise headline placement, clean studio light, and balanced negative space.
Using "nano-banana-2". Create an image about a public event from this week.
Expected outcome:
Recommended setup: enable web grounding only for this current-event request. Confirm that the prompt may be sent to RunComfy with added latency and cost.
Security Audit
High RiskMost static findings are Markdown formatting, schema field names, documentation links, or sample prompts rather than executable code. Confirmed risks are the intended RunComfy CLI execution, setup/login commands, and disclosed outbound RunComfy API behavior. I also found an installation-source mismatch that could direct users to install a different repository.
Confirmed security concerns (1)
Capability review items (11)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (40)
🌐 Network access (4)
📁 Filesystem access (3)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/doany-ai-nano-banana-2/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/doany-ai-nano-banana-2?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/doany-ai-nano-banana-2?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/doany-ai-nano-banana-2/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/doany-ai-nano-banana-2.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
doany-ai. (2026). nano-banana-2 security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/doany-ai-nano-banana-2/audits/4BibTeX citation
@techreport{doany-ai-doany-ai-nano-banana-2-2026,
author = {doany-ai},
title = {nano-banana-2 security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/doany-ai-nano-banana-2/audits/4},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "nano-banana-2 security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
- name: "doany-ai"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/doany-ai-nano-banana-2/audits/4"
identifiers:
- type: other
value: "skillstore:doany-ai-nano-banana-2:audit:4"
description: "Skillstore immutable audit report identifier"
Compare variants
8 installable variantsEach author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.
Why this variant is first
inferen-sh-nano-banana-2
2026-08-21
agentspace-so-nano-banana-2
2026-08-21
toolshell-nano-banana-2
2026-08-21
inference-shell-nano-banana-2
2026-08-21
tul-sh-nano-banana-2
2026-08-21
doany-ai-nano-banana-2
2026-08-21
inference-sh-9-nano-banana-2
2026-08-21
runcomfy-com-nano-banana-2
2026-08-21
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Create social thumbnail batches
Generate four fast draft options with platform-specific ratios before selecting a final direction.
Draft branded text visuals
Use quoted text and placement guidance to improve posters, product cards, and launch graphics.
Explore current-event imagery
Enable web grounding only when prompts need current entities, events, or public context.
Try These Prompts
Generate one Nano Banana 2 image for this concept: [describe subject, setting, mood, and style]. Use the default 1K square output.
Create four Nano Banana 2 drafts for [campaign or asset]. Use [aspect ratio], 0.5K resolution, and keep the subject consistent across variants.
Generate a [poster/card/banner] with the exact text "[text]" placed [position]. Specify font style, contrast, subject, background, and aspect ratio.
Refine this selected Nano Banana 2 direction using seed [number], [2K or 4K] resolution, [aspect ratio], and these changes: [specific visual changes].
Best Practices
- Front-load the subject, action, environment, style, and camera details in that order.
- Quote exact in-image text and specify its placement, contrast, and font style.
- Lock the seed when refining a selected composition across small prompt changes.
Avoid
- Do not use this endpoint for image edits that must preserve a real subject identity.
- Do not combine many conflicting style anchors in one prompt.
- Do not enable web search unless the image needs current real-world context.