nano-banana-prompts

Generate optimized prompts for Gemini 2.5 Flash Image (Nano Banana). Use for image generation, crafting photo prompts, art styles, or multi-turn editing workflows with best practices.

1.03x

Quality

78%

Does it follow best practices?

Impact

97%

1.03x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/nano-banana-prompts/skills/nano-banana-prompts/SKILL.md

Quality

Content

77%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable content with excellent templates, model-specific constraints, and well-sequenced workflows with feedback loops. The weaknesses are redundancy — three overlapping checklists and three overlapping pitfall/troubleshooting sections — and a monolithic single-file structure that inlines material better suited to one-level-deep reference files.

Suggestions

Consolidate the three overlapping quality-control passes (Quick Prompt Checklist, Prompt Validation, and Text Rendering Validation) into a single validation section, and merge Common Pitfalls, Common Issues & Solutions, and the Troubleshooting Guide table into one troubleshooting reference.

Move the per-category deep dives (Product Photography, Landscape Photography, Abstract Art, and the long validation/troubleshooting material) into references/ files (e.g., references/templates.md, references/troubleshooting.md), keeping SKILL.md as a concise overview with clearly signaled one-level-deep links.

Either reference assets/nano-banana.png in the body where it is useful, or remove it, since it is currently an orphaned placeholder in the bundle.

Dimension	Reasoning	Score
Conciseness	Templates, camera specs, and worked examples earn their tokens, but the document passes over the same material repeatedly: 'Quick Prompt Checklist' duplicates 'Prompt Validation → Content Completeness', 'Text Rendering' best practices duplicate 'Text Rendering Validation', and 'Common Pitfalls', 'Common Issues & Solutions', and the 'Troubleshooting Guide' table cover the same ground three times. Mostly efficient but could be tightened by consolidating these overlapping sections.	3 / 5
Actionability	Copy-paste-ready templates for every category (photography, art, product, landscape, abstract), fully worked example prompts, and specific model constraints ('Keep under 25 characters total', '85mm f/2.0', '50-100 words: RECOMMENDED') provide complete executable guidance.	5 / 5
Workflow Clarity	Multi-step processes are clearly sequenced with validation checkpoints and feedback loops: the Testing section's 6 evaluation criteria, the Refinement Workflow (identify gap → targeted one-change edit → refine for future use), and the Comparison Test (Versions A/B/C). The one-change-per-turn rule is enforced with an explicit good/bad example.	5 / 5
Progressive Disclosure	Headers keep the 410-line single file navigable, but content that belongs in separate reference files (per-category templates, validation checklists, the troubleshooting table) is fully inlined, and no references/ files exist at all — the only bundle file is an unreferenced 14-byte placeholder PNG. This matches 'some structure but content that should be separate is inline'.	3 / 5
	Total	16 / 20 Passed

Description

78%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with an explicit 'Use for...' trigger clause, concrete trigger phrases, and a clearly named target model. The what-clause covers a single action rather than a fuller capability list, and a few natural synonyms (text-to-image, image editing) would round out trigger coverage.

Dimension	Reasoning	Score
Specificity	The description states one concrete action — 'Generate optimized prompts for Gemini 2.5 Flash Image (Nano Banana)' — naming the domain precisely but not enumerating multiple distinct capabilities (e.g., editing workflows or style guidance) as separate actions in the what-clause, matching the '1-2 concrete actions, but not comprehensive' anchor.	3 / 5
Completeness	It explicitly answers both what ('Generate optimized prompts for Gemini 2.5 Flash Image (Nano Banana)') and when ('Use for image generation, crafting photo prompts, art styles, or multi-turn editing workflows') with concrete trigger phrases in third person, matching the top anchor.	5 / 5
Trigger Term Quality	Natural triggers like 'image generation', 'crafting photo prompts', 'art styles', and 'multi-turn editing workflows' plus the model names 'Gemini 2.5 Flash Image' and 'Nano Banana' give good coverage, but common variations such as 'text-to-image', 'generate images', or 'image editing' are missing.	4 / 5
Distinctiveness Conflict Risk	The named model gives a clear niche with distinct triggers, but broad terms like 'image generation' and 'art styles' create minor overlap risk with any general image-generation skill; mostly distinct rather than minimal-conflict.	4 / 5
	Total	16 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: secondsky/claude-skills
Path: plugins/nano-banana-prompts/skills/nano-banana-prompts/SKILL.md
Commit: 8837836

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.