nano-banana-pro

Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/nano-banana-pro/SKILL.md

Quality

Content

96%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 body is an exemplary lean, executable reference: commands are copy-paste ready and verified against the bundled script's actual CLI, and every note is operationally relevant. The only shortcoming is cosmetic — section labels are plain text rather than markdown headings, which slightly weakens navigation.

Dimension	Reasoning	Score
Conciseness	The ~30-line body is lean and assumes Claude's competence: three bare commands, API-key configuration, and four operational notes ("Resolutions: 1K (default), 2K, 4K", "Do not read the image back; report the saved path only"), with no concept explanations or padding. Every token earns its place, matching the anchor-5 example's economy.	5 / 5
Actionability	All three usage modes (generate, single-image edit, multi-image composition) are given as copy-paste-ready `uv run` commands, and the flags (--prompt, --filename, -i, --resolution) match the actual bundled script's CLI exactly. The examples cover the common cases, matching the 'fully executable' anchor; not scored lower because no key execution detail is missing.	5 / 5
Workflow Clarity	This is a simple single-purpose skill where the single action (run generate_image.py with a prompt and filename) is unambiguous, and generation is neither destructive nor batch, so the simple-skill exception applies. The notes even specify the post-step behavior (report the saved path, do not read the image back).	5 / 5
Progressive Disclosure	The body is under 50 lines, needs no external reference files, and its single bundle reference ({baseDir}/scripts/generate_image.py) is a real, correctly-pathed, one-level-deep script. Falls short of 5 only because section labels ("Generate", "Multi-image composition (up to 14 images)", "Notes") are bare text lines rather than markdown headers, a minor organization gap matching anchor 4.	4 / 5
	Total	19 / 20 Passed

Description

53%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 concise and names a concrete, distinctive capability tied to a specific model, but it omits any explicit trigger guidance and has thin keyword coverage. Adding a "Use when..." clause with natural synonyms (create images, photo editing, nano banana) and mentioning additional capabilities like multi-image composition would materially raise the score.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to generate, create, or edit images, photos, or pictures, or mentions Nano Banana Pro / Gemini 3 Pro Image."

Broaden keyword coverage with natural synonyms users would say: "create", "make", "photo", "picture", "image composition", "AI image generation".

Mention distinctive capabilities from the body (multi-image composition up to 14 images, 1K/2K/4K resolutions) so the 'what' is more comprehensive and the skill is harder to confuse with generic image tools.

Dimension	Reasoning	Score
Specificity	"Generate or edit images" names the domain plus two concrete actions, matching the anchor for 1-2 concrete actions without comprehensive coverage (multi-image composition, resolutions, and editing modes are absent). Not 4 because it does not list several specific actions; not 2 because the actions are concrete rather than generic.	3 / 5
Completeness	The description gives a clear 'what' ("Generate or edit images via Gemini 3 Pro Image") but has no "Use when..." clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. Not 4 because the 'when' is entirely absent rather than merely imprecise.	3 / 5
Trigger Term Quality	"Generate", "edit", "images", "Gemini 3 Pro Image", and "Nano Banana Pro" are relevant keywords, but common variations users naturally say ("create", "make", "photo", "picture") are missing. Fits the 'some relevant keywords but missing common variations' anchor rather than the good-coverage anchor.	3 / 5
Distinctiveness Conflict Risk	Naming the specific model ("Gemini 3 Pro Image (Nano Banana Pro)") carves out a clear niche, but the generic "generate or edit images" phrase overlaps with any other image-generation or editing skill. Mostly distinct with minor overlap risk, matching anchor 4; not 5 because the trigger surface is not uniquely staked out.	4 / 5
	Total	13 / 20 Passed

Validation

81%

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 — 13 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: Bitterbot-AI/bitterbot-desktop
Path: skills/nano-banana-pro/SKILL.md
Commit: 1c7dc1b

Reviewed: 20 hours 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.