gpt-image-skill

Generate or edit images using OpenAI GPT Image API (gpt-image-2, gpt-image-1, etc). Use ONLY when the user explicitly names OpenAI or GPT as the provider: "gpt image", "openai image", "generate image with openai", "用 openai 画图", "用 GPT 生成图片". For generic image requests without a provider, use nanobanana-skill instead. Do NOT use for diagrams (架构图/流程图) — draw those with Mermaid or code.

Quality

85%

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

Quality

Content

75%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 highly actionable, well-structured CLI-wrapper skill with copy-paste executable examples and clear workflows. The main weakness is mild redundancy across the ask-user, Available Options, and Examples sections, plus some generic Best Practices advice that could be trimmed.

Suggestions

Consolidate the option list: avoid repeating size/quality/model/format values across the ask-user step, Available Options, and Examples — keep one canonical reference and have examples show only the distinctive flag per case.

Trim generic Best Practices tips (e.g. "Be descriptive in prompts") in favor of GPT-Image-specific guidance that Claude would not already infer.

Tighten workflow_clarity by integrating the Error Handling checklist as an explicit feedback step (e.g. "If the script fails: check OPENAI_API_KEY, then re-run") rather than leaving it as a separate trailing section.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with executable commands, but the ask-user option list, Available Options reference, and Examples overlap, and Best Practices offers generic advice Claude already knows ("Be descriptive in prompts — include style, mood, colors").	3 / 5
Actionability	Fully executable, copy-paste-ready bash commands with concrete flags cover the common cases (generation, editing, multi-image, transparent background, model selection).	5 / 5
Workflow Clarity	Clear sequenced workflow (ask user → run script → show path) with a dedicated Error Handling section for recovery, though checkpoints are not explicitly wired into the step flow as feedback loops.	4 / 5
Progressive Disclosure	Well-organized sections (Requirements, Instructions, Options, Examples, Error Handling, Best Practices) with no nested references; the executable script is the only external path and is clearly signaled, though the ~180-line body could offload some option reference to a separate file.	4 / 5
	Total	16 / 20 Passed

Description

95%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, well-gated description that clearly answers what and when, with rich natural trigger terms including bilingual variants and explicit boundaries against competing skills. The only minor gap is that the action list (generate/edit) is somewhat compact versus the most comprehensive examples.

Dimension	Reasoning	Score
Specificity	"Generate or edit images using OpenAI GPT Image API" names two concrete actions (generate, edit) plus model variants, which is specific but slightly short of comprehensive coverage.	4 / 5
Completeness	Explicitly states what it does ("Generate or edit images using OpenAI GPT Image API") and when to use it ("Use ONLY when the user explicitly names OpenAI or GPT as the provider"), with concrete trigger phrases and exclusions.	5 / 5
Trigger Term Quality	Includes natural trigger phrases users would say — "gpt image", "openai image", "generate image with openai" — plus bilingual synonyms ("用 openai 画图", "用 GPT 生成图片"), giving comprehensive coverage.	5 / 5
Distinctiveness Conflict Risk	The provider-name gate ("Use ONLY when the user explicitly names OpenAI or GPT"), redirect of generic requests to nanobanana-skill, and diagram exclusion to Mermaid create a clear niche with minimal conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	15 / 16 Passed

Repository: feiskyer/claude-code-settings
Path: skills/gpt-image-skill/SKILL.md
Commit: 6e19e61

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.