imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG/vector/code-native assets, extending an established icon or logo system, or building the visual directly in HTML/CSS/canvas.

Quality

90%

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

88%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-structured, highly actionable skill body with a clear workflow, validation feedback loops, and good reference externalization. Its main weaknesses are mild cross-section rule repetition and a couple of reference-dense tables inlined into SKILL.md rather than delegated to bundle files.

Suggestions

DRY the recurring rules: state CLI-fallback gating, transparent-background handling, and the save-path policy once in their canonical sections and reference them from the workflow/best-practices sections instead of restating.

Move the gpt-image-2 size/quality table and possibly the use-case taxonomy into a reference file (e.g. references/image-api.md or a taxonomy reference), keeping only the most common sizes and the slug list summary in SKILL.md.

Consider adding file-extension triggers (.png/.jpg) to the description to push trigger-term quality to the top anchor.

Dimension	Reasoning	Score
Conciseness	The body is dense with actionable, non-obvious specifics (no padding explaining what a PNG is or how libraries work), but several rule sets are restated across sections — CLI-fallback gating, transparent handling, and the save-path policy each recur 3–4 times — which could be DRY'd.	4 / 5
Actionability	Provides a copy-paste-ready shared prompt schema, two complete worked examples (generation and edit with verbatim specs), concrete commands ("uv pip install openai"), and specific size/quality tables covering the common cases.	5 / 5
Workflow Clarity	An 18-step workflow has explicit validation (step 12 inspects subject, style, composition, text accuracy, and invariants), a feedback loop (step 13: iterate with a single change then re-check), and a decision-tree checklist, satisfying the validation requirement for batch-capable operations.	5 / 5
Progressive Disclosure	A clear "Reference map" labels six one-level-deep references and externalizes the bulk (sample prompts, full API reference, CLI usage, network troubleshooting), but the inlined gpt-image-2 size/quality table and the 19-slug use-case taxonomy are reference-dense content that could live in a reference file.	4 / 5
	Total	18 / 20 Passed

Description

92%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 capabilities, gives concrete trigger terms and use-when guidance, and draws a sharp boundary against vector/code-native alternatives. The only gap is the absence of file-extension triggers, which keeps trigger-term quality just below the top anchor.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ("Generate or edit raster images", "create a brand-new image, transform an existing image, or derive visual variants") plus a comprehensive set of asset types (photos, illustrations, textures, sprites, mockups, transparent-background cutouts).	5 / 5
Completeness	Explicitly answers "what" (generate/edit raster bitmap visuals) and "when" ("Use when Codex should create a brand-new image, transform an existing image, or derive visual variants") with concrete trigger phrases, plus an explicit "Do not use when" boundary.	5 / 5
Trigger Term Quality	Strong natural keyword coverage (photos, illustrations, textures, sprites, mockups, image, transparent background, cutouts), but no file extensions (.png/.jpg) or common synonyms like "picture", which the anchor-5 examples treat as a hallmark of comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	Clear AI-generated bitmap niche with an explicit boundary against SVG/vector/code-native assets, icon/logo systems, and HTML/CSS/canvas, minimizing overlap with related skills.	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
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: openai/codex
Path: codex-rs/skills/src/assets/samples/imagegen/SKILL.md
Commit: 78245b4

Reviewed: 5 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.