nano-banana-pro-openrouter

Generate or edit images via OpenRouter with the Gemini 3 Pro Image model. Use for prompt-only image generation, image edits, and multi-image compositing; supports 1K/2K/4K output.

3.10x

Quality

86%

Does it follow best practices?

Impact

90%

3.10x

Average score across 3 eval scenarios

Securityby

Critical

Do not install without reviewing

Quality

Content

93%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.

The skill body is lean, executable, and well-structured, with copy-paste commands covering all three generation modes and a concrete troubleshooting table. The only notable gap is the absence of an explicit validation/recovery feedback loop for the (mildly batch) multi-image case.

Suggestions

Add a short validate-then-proceed step for multi-image composition, e.g. confirm expected input count before invoking and verify saved outputs via the printed 'MEDIA: <path>' lines.

Optionally surface the SYSTEM_TEMPLATE customization option earlier (near the examples) so users discover it without reading to the end.

Dimension	Reasoning	Score
Conciseness	The body is lean and efficient with no concept-padding or restatement of what Claude already knows; each section (overview, modes, resolution, troubleshooting) earns its place. It is not lower because nothing is padded or over-explained.	5 / 5
Actionability	Provides copy-paste-ready 'uv run' commands with real flags (--prompt, --input-image, --filename, --resolution) and a troubleshooting table with concrete fixes for each blocker. It is not lower because the common cases are fully executable.	5 / 5
Workflow Clarity	Each generation mode has a clearly sequenced example with flags, plus retry guidance for transient errors; checkpoints are mostly present. It is not a 5 because there is no explicit validate-then-proceed feedback loop, and not a 3 because the sequence and error handling are concrete.	4 / 5
Progressive Disclosure	Well-organized sections with one-level-deep references to real bundle files (scripts/generate_image.py and assets/SYSTEM_TEMPLATE), both verified to exist, and no nested reference chains. It is not lower because structure and navigation are clear.	5 / 5
	Total	19 / 20 Passed

Description

80%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.

The description is strong, explicitly covering both what the skill does and when to use it with a concrete trigger clause and distinct niche. Its main weakness is trigger-term breadth, leaning on technical model/provider names rather than the everyday terms and file extensions users would naturally mention.

Suggestions

Add natural synonyms and file extensions users would say, e.g. 'image generation, image editing, .png/.jpg edits, multi-image composition'.

Reduce reliance on technical jargon ('Gemini 3 Pro Image', 'OpenRouter') as the primary trigger terms in favor of user-facing language.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Generate or edit images', 'prompt-only image generation, image edits, and multi-image compositing', '1K/2K/4K output') with only minor gaps such as omitted output file formats. It is not a 5 because coverage stops short of fully comprehensive action enumeration.	4 / 5
Completeness	Clearly answers 'what' (generate/edit images via OpenRouter with Gemini 3 Pro Image) and 'when' via the explicit 'Use for prompt-only image generation, image edits, and multi-image compositing' trigger clause with concrete phrases. Re-reading anchors, this matches the explicit what-and-when anchor better than 4.	5 / 5
Trigger Term Quality	Includes relevant natural terms ('image generation', 'image edits', 'multi-image compositing') but relies on technical jargon ('OpenRouter', 'Gemini 3 Pro Image') and omits common synonyms and file extensions (.png, .jpg) that users would naturally say. It is not a 4 because a few common natural terms are still missing.	3 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (OpenRouter Gemini 3 Pro Image generation) with distinct triggers and minimal overlap risk with other skills. It is not lower because the domain and tooling are specific enough to avoid misfiring.	5 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: github/awesome-copilot
Path: skills/nano-banana-pro-openrouter/SKILL.md
Commit: 406c31f

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.