muapi-nano-banana

Reasoning-driven image generation using structured creative briefs (Gemini 3 style) — generates high-fidelity images via muapi.ai with logic-based prompting

Quality

50%

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

Fix and improve this skill with Tessl

tessl review fix ./.opencode/skills/muapi-nano-banana/SKILL.md

Quality

Content

50%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill body is well-structured and gives concrete prompting guidance plus a bash example, but it references scripts that are missing from the bundle and carries conceptual padding. Every dimension lands at the middle of the scale.

Suggestions

Ship the referenced scripts/generate-nano-art.sh (or remove the broken references) so the bash example is genuinely executable and progressive-disclosure references resolve to real files.

Tighten the workflow: explicitly show how the Protocol's rewritten Reasoning Brief is passed into the script invocation, and make Step 1 concrete instead of "Provide the agent with a subject and a specific scenario."

Trim the Gemini 3 backstory and "moves beyond keyword stuffing" prose to reduce token overhead and respect Claude's existing knowledge of prompting.

Dimension	Reasoning	Score
Conciseness	The body is reasonably sectioned but carries marketing/conceptual padding ("Based on the advanced prompting architecture of Google's Gemini 3", "moves beyond keyword stuffing") and instructive philosophy Claude largely already knows. Not level 3 because not every token earns its place; not level 1 because it avoids a beginner-explainer wall of text.	2 / 3
Actionability	It provides a concrete bash example with specific flags (--subject, --action, --context, --style) and example values, but the referenced scripts/generate-nano-art.sh is absent from the bundle, so the example is not actually copy-paste executable. Not level 3 because the executable target is missing; not level 1 because concrete flags and examples are present rather than abstraction.	2 / 3
Workflow Clarity	Two steps are listed (define creative logic, invoke the script), but Step 1 is vague ("Provide the agent with a subject and a specific scenario") and the bridge from the Protocol's Reasoning Brief to script invocation is not crisply sequenced, with no validation checkpoints. Not level 3 because the sequence is loose and Step 1 is underspecified; not level 1 because steps are present.	2 / 3
Progressive Disclosure	Sections are clearly organized, but the body references external scripts (scripts/generate-nano-art.sh, core/media/generate-image.sh) that do not exist in the bundle, so the references are broken. Not level 3 because references are unfulfilled; not level 1 because there is no monolithic wall or deeply nested reference chain.	2 / 3
	Total	8 / 12 Passed

Description

50%

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 states a clear, third-person capability tied to a specific provider, but it is jargon-heavy and omits any explicit "Use when" trigger guidance. It lands at the middle of the scale across all dimensions.

Suggestions

Add an explicit 'Use when...' clause with natural user triggers, e.g. 'Use when generating AI images, creating artwork, or when the user asks to generate or edit a picture.'

Swap jargon-heavy phrasing ("structured creative briefs", "logic-based prompting") for common terms users actually say ("AI image generation", "generate a picture") to improve trigger-term coverage.

Name additional concrete actions (e.g., render legible text in images, maintain subject consistency) to lift specificity toward multiple distinct capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain and a concrete action ("generates high-fidelity images via muapi.ai") but lists only one action surrounded by abstract qualifiers like "structured creative briefs" and "logic-based prompting". It is not level 3 because it lacks multiple distinct concrete actions, and not level 1 because concrete actions and the domain are explicitly stated.	2 / 3
Completeness	It clearly answers what the skill does ("generates high-fidelity images via muapi.ai") but provides no "Use when..." clause or equivalent trigger for when to use it, capping completeness at 2 per the judging guidelines. Not level 3 because the when is entirely missing; not level 1 because the what is explicit.	2 / 3
Trigger Term Quality	"image generation" and "images" are natural terms, but the description leans on jargon ("structured creative briefs", "logic-based prompting", "Gemini 3 style") and omits common phrasings like "AI art" or "create a picture". Not level 3 due to incomplete natural coverage; not level 1 because some genuine user terms appear.	2 / 3
Distinctiveness Conflict Risk	The muapi.ai / nano-banana niche is fairly specific, but there are no explicit trigger terms to distinguish invocation from a generic image-generation skill, so overlap remains possible. Not level 3 because distinct triggers are absent; not level 1 because the niche itself is not generic.	2 / 3
	Total	8 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: SamurAIGPT/Generative-Media-Skills
Path: .opencode/skills/muapi-nano-banana/SKILL.md
Commit: 71242e2

Reviewed: about 18 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.