nano-banana-pro
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).
70
60%
Does it follow best practices?
Impact
94%
7.23xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./openclaw/skills/nano-banana-pro/SKILL.mdQuality
Discovery
32%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 concise but too sparse. It identifies the core capability (image generation/editing) and the specific model, but lacks a 'Use when...' clause, detailed action list, and natural trigger terms users would employ. Adding explicit trigger guidance and more user-facing keywords would significantly improve skill selection accuracy.
Suggestions
Add a 'Use when...' clause specifying triggers like 'Use when the user asks to create, generate, edit, or modify images, pictures, or photos using Gemini or Nano Banana Pro.'
Include more natural trigger terms users would say, such as 'create image', 'picture', 'photo', 'illustration', 'AI-generated image', 'image editing'.
List more specific capabilities beyond 'generate or edit', such as 'apply style changes, add or remove objects, upscale, or transform existing images' to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (image generation/editing) and two actions (generate, edit), but doesn't list more specific capabilities like style transfer, inpainting, resizing, format conversion, etc. | 2 / 3 |
Completeness | Describes what it does (generate or edit images via a specific model) but completely lacks a 'Use when...' clause or any explicit trigger guidance, which per the rubric should cap completeness at 2, and since the 'what' is also thin, this lands at 1. | 1 / 3 |
Trigger Term Quality | Includes 'generate' and 'edit images' which are natural terms, plus the model name 'Gemini 3 Pro Image' and 'Nano Banana Pro' for specific lookups. However, it misses common user terms like 'create image', 'picture', 'photo', 'illustration', 'draw', or 'AI image'. | 2 / 3 |
Distinctiveness Conflict Risk | The specific model name 'Gemini 3 Pro Image (Nano Banana Pro)' adds some distinctiveness, but 'generate or edit images' is broad enough to overlap with other image-related skills like DALL-E or Stable Diffusion skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a lean, well-structured skill that provides clear, executable commands for all supported use cases. Its main weakness is the lack of any error handling or validation guidance—what to do if the API key is invalid, if the input image doesn't exist, or if the generation fails. The conciseness and actionability are excellent, respecting Claude's intelligence while providing all necessary specifics.
Suggestions
Add a brief note on error handling: what common failures look like (e.g., missing API key, invalid image path) and how to recover.
Consider adding a validation step such as checking the output file exists after generation, especially for multi-image composition which may be more failure-prone.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every line serves a purpose. No unnecessary explanations of what Gemini is or how image generation works. The notes section is tightly packed with only non-obvious, actionable details. | 3 / 3 |
Actionability | Provides fully executable, copy-paste-ready bash commands for all three use cases (generate, edit, multi-image). API key configuration is specific with exact config paths. Concrete resolution options and filename conventions are given. | 3 / 3 |
Workflow Clarity | The three use cases are clearly presented with distinct commands, but there's no validation or error handling guidance. What happens if the API key is missing? What if the script fails? For a skill that calls an external API and writes files, some feedback loop or error checking would be valuable. | 2 / 3 |
Progressive Disclosure | For a simple, single-purpose skill under 50 lines, the content is well-organized with clear sections (Generate, Edit, Multi-image, API key, Notes). The script itself is referenced rather than inlined, which is appropriate progressive disclosure. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 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 |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 8 / 11 Passed | |
- Repository
- trpc-group/trpc-agent-go
- Commit
9f4b734
- 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.