nano-banana-pro
Generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image) API. Use when the user asks to generate, create, edit, modify, change, alter, or update images. Also use when user references an existing image file and asks to modify it in any way (e.g., "modify this image", "change the background", "replace X with Y"). Supports both text-to-image generation and image-to-image editing with configurable resolution (1K default, 2K, or 4K for high resolution). DO NOT read the image file first - use this skill directly with the --input-image parameter.
88
82%
Does it follow best practices?
Impact
96%
2.82xAverage score across 3 eval scenarios
Risky
Do not use 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 skill description that excels across all dimensions. It provides specific capabilities (text-to-image, image-to-image, resolution options), comprehensive trigger terms covering natural user language, an explicit 'Use when...' clause with diverse scenarios, and clear distinctiveness through its specific API and domain focus. The inclusion of operational guidance ('DO NOT read the image file first') adds practical value for correct skill execution.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: generate images, edit images, text-to-image generation, image-to-image editing, configurable resolution options (1K, 2K, 4K). Also mentions specific technical details like the --input-image parameter. | 3 / 3 |
Completeness | Clearly answers both 'what' (generate and edit images using Gemini API with configurable resolution) and 'when' (explicit 'Use when...' clause with multiple trigger scenarios including both generation and editing use cases). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would say: 'generate', 'create', 'edit', 'modify', 'change', 'alter', 'update images', 'change the background', 'replace X with Y'. These are highly natural phrases users would actually use. | 3 / 3 |
Distinctiveness Conflict Risk | Clearly scoped to image generation/editing via a specific API (Nano Banana Pro / Gemini 3 Pro Image). The combination of the specific API, image-focused triggers, and the operational note about --input-image makes it highly distinct from other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with clear executable commands and good coverage of both generation and editing workflows. Its main weaknesses are moderate verbosity (especially in filename generation and resolution mapping sections) and the lack of error handling/validation steps for API calls that could fail. The structure is functional but could benefit from tighter organization.
Suggestions
Add error handling guidance: what to do when the API call fails (e.g., rate limits, invalid key, network errors) to improve workflow clarity.
Trim the filename generation section - the pattern and 1-2 examples suffice; Claude can infer the rest from the format description.
Consider condensing the resolution mapping into a simpler table rather than listing both the options and the user-request mappings separately.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient but includes some sections that could be tightened. The filename generation section with multiple examples is somewhat verbose, and the API key resolution order is something Claude could infer. The resolution mapping table adds value but the 'Map user requests' subsection is borderline unnecessary. | 2 / 3 |
Actionability | Provides fully executable, copy-paste ready bash commands with all parameters clearly shown. The examples section gives concrete invocations for both generation and editing use cases with realistic arguments. | 3 / 3 |
Workflow Clarity | The image editing section lists steps but lacks validation checkpoints. There's no guidance on what to do if the script fails, no error handling workflow, and no verification step after generation. For a skill that calls an external API and writes files, a feedback loop for failures would be appropriate. | 2 / 3 |
Progressive Disclosure | Content is reasonably organized with clear section headers, but everything is inline in a single file. The filename generation rules and resolution mapping could potentially be more concise. For a skill of this length (~80 lines of content), the organization is adequate but the sections could be better prioritized with quick-start info first and details later. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- intellectronica/agent-skills
- Commit
9b0e00a
- 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.