nano-banana-pro

Generate or edit a single image via OpenRouter (google/gemini-3.1-flash-image-preview by default). Accepts a text prompt and optional --input-image for image-to-image editing. Trigger when the user asks for an AI image, illustration, concept art, product render, or wants to modify an existing image.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

67%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.

A well-organized, mostly lean body that gives concrete inputs, output/receipt semantics, and failure handling for a single-image generation skill. The main weakness is an over-detailed Auth section that explains script internals Claude does not need to drive the skill.

Suggestions

Trim the Auth section to the key resolution order and move URL-hardening internals (userinfo/fragment/port rejection, redirect handling, opensquilla.toml non-parsing) into a references/auth.md file or a script docstring.

Add one copy-paste-ready example invocation in the body (e.g. a full `python scripts/generate_image.py --prompt ... --filename ... --aspect-ratio 9:16` line) covering the common generate case.

Surface the bundled script path explicitly in the body ("implementation: scripts/generate_image.py") so the one-level-deep reference to the bundle is clearly signaled.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with a well-structured inputs table and cost/latency section, but the Auth section spends many lines describing script internals (URL userinfo/fragment/port rejection, redirect handling, opensquilla.toml parsing, profile-pool parking) that could be tightened or moved to a reference file.	3 / 5
Actionability	The inputs table gives concrete keys, defaults, and allowed values, and edit mode is documented via "invoke the script directly with --input-image PATH"; the gap is that no complete worked invocation example appears in the body to copy-paste.	4 / 5
Workflow Clarity	A single-action skill whose action is unambiguous, with described in-script validation (MIME/format/dimension/integrity checks before receipt write) and explicit paid-request retry semantics; not a 5 because there is no explicit validate→fix→retry loop a caller follows.	4 / 5
Progressive Disclosure	Clear sectioned overview (Inputs, Auth, Output, Cost/latency, Common failures) with the one bundle file (scripts/generate_image.py) referenced via the entrypoint; the dense Auth section is inline rather than split to a reference, a minor organization gap.	4 / 5
	Total	15 / 20 Passed

Description

100%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.

A strong, third-person description that crisply states capability, names the default model/transport, and gives explicit natural-language triggers. It cleanly answers both what and when with no fluff.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — "Generate or edit a single image", "image-to-image editing", "Accepts a text prompt and optional --input-image" — plus the default model and transport (OpenRouter), giving comprehensive coverage of this single-image domain.	5 / 5
Completeness	Explicitly answers both what ("Generate or edit a single image via OpenRouter...") and when ("Trigger when the user asks for an AI image, illustration, concept art, product render...") with concrete trigger phrases.	5 / 5
Trigger Term Quality	"Trigger when the user asks for an AI image, illustration, concept art, product render, or wants to modify an existing image" gives comprehensive natural-term coverage with synonyms a user would actually say; no file-extension trigger applies to image generation.	5 / 5
Distinctiveness Conflict Risk	Carves a clear niche — single-image generation/editing via OpenRouter — with distinct triggers unlikely to fire for unrelated skills.	5 / 5
	Total	20 / 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: TokenRhythm/opensquilla
Path: src/opensquilla/skills/bundled/nano-banana-pro/SKILL.md
Commit: 94ac35e

Reviewed: 4 days 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.