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.
90
89%
Does it follow best practices?
Impact
90%
3.10xAverage score across 3 eval scenarios
Critical
Do not install without reviewing
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong, well-crafted description that clearly identifies the specific tool chain (OpenRouter + Gemini 3 Pro Image), lists concrete capabilities (generation, editing, compositing), and provides explicit 'Use for...' trigger guidance. The inclusion of resolution details (1K/2K/4K) adds useful specificity. It uses proper third-person voice throughout.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'prompt-only image generation', 'image edits', 'multi-image compositing', and specifies output resolutions '1K/2K/4K output'. Also names the specific model (Gemini 3 Pro) and platform (OpenRouter). | 3 / 3 |
Completeness | Clearly answers both 'what' (generate/edit images via OpenRouter with Gemini 3 Pro, supporting various resolutions) and 'when' ('Use for prompt-only image generation, image edits, and multi-image compositing'), with explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms: 'generate images', 'edit images', 'image generation', 'image edits', 'compositing', 'OpenRouter', 'Gemini'. Users asking about image generation or editing would naturally use these terms. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to specifying the exact platform (OpenRouter), model (Gemini 3 Pro Image), and specific capabilities (multi-image compositing, resolution options). Unlikely to conflict with other image skills unless another skill targets the same model. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
79%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted, concise skill that provides immediately actionable commands for three distinct use cases. Its main strengths are token efficiency and concrete examples. The primary weakness is the lack of explicit validation steps after image generation (e.g., confirming the output file was created successfully) and the workflow could benefit from a clearer sequential structure for error recovery.
Suggestions
Add a brief validation step after generation, e.g., 'Verify the output exists and check the MEDIA: line in stdout before reporting success to the user.'
Consider linking to a separate REFERENCE.md or similar for detailed script arguments and SYSTEM_TEMPLATE format documentation.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It doesn't explain what image generation is or how OpenRouter works — it jumps straight to executable commands. Every section earns its place, including the troubleshooting table which provides genuinely useful non-obvious information. | 3 / 3 |
Actionability | All three usage patterns (prompt-only, single edit, multi-image) are shown with concrete, copy-paste-ready commands. The troubleshooting table provides specific resolutions with exact commands for different platforms. The resolution flag and constraints are clearly specified. | 3 / 3 |
Workflow Clarity | The skill presents clear individual commands but lacks an explicit workflow sequence for multi-step operations. There's no validation checkpoint (e.g., verify the output image exists or check file size) after generation. The retry guidance for transient errors is good but the overall process doesn't have a validate-then-proceed structure. | 2 / 3 |
Progressive Disclosure | The content is well-organized with clear sections, but the troubleshooting table and behavior constraints are inline rather than referenced from a separate file. For a skill of this size (~60 lines of content), this is borderline acceptable, but the mention of `assets/SYSTEM_TEMPLATE` and `scripts/generate_image.py` without linking to further documentation about their internals is a minor gap. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 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 | 9 / 11 Passed | |
- Repository
- github/awesome-copilot
- Commit
4020587
- Reviewed
Table of Contents
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.