gpt-image-1-5
Generate and edit images using OpenAI's GPT Image 1.5 model. Use when the user asks to generate, create, edit, modify, change, alter, or update images. Also use when user references an existing image file and asks to modify it in any way (e.g., "modify this image", "change the background", "replace X with Y"). Supports text-to-image generation and image editing with optional mask. DO NOT read the image file first - use this skill directly with the --input-image parameter.
87
82%
Does it follow best practices?
Impact
94%
2.29xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that clearly communicates what the skill does (image generation and editing via GPT Image 1.5), when to use it (with extensive trigger terms and example phrases), and how it's distinct from other skills. The inclusion of concrete user phrases and the operational note about not reading image files first adds practical value. Minor note: the version '1.5' may not be a real model name, but that doesn't affect the description quality.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: generate images, edit images, text-to-image generation, image editing with optional mask. Also specifies the model (GPT Image 1.5) and a specific parameter (--input-image). | 3 / 3 |
Completeness | Clearly answers both 'what' (generate and edit images using GPT Image 1.5, text-to-image, image editing with mask) and 'when' (explicit 'Use when...' clause with multiple trigger scenarios including referencing existing image files). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms: 'generate', 'create', 'edit', 'modify', 'change', 'alter', 'update images', plus concrete user phrases like 'modify this image', 'change the background', 'replace X with Y'. These are terms users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Clearly scoped to image generation/editing via a specific model (GPT Image 1.5). The mention of the specific model and --input-image parameter makes it highly distinct from other potential skills like image analysis or general file manipulation. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with clear executable commands and good parameter documentation. Its main weaknesses are some redundancy between the Usage and Examples sections, and the lack of error handling/validation guidance for API failures or invalid inputs. The content structure is reasonable but slightly verbose for what is essentially a single-script wrapper.
Suggestions
Remove the duplicate Examples section at the bottom since the Usage section already contains complete command examples, or consolidate into a single section
Add brief error handling guidance: what to do if the API key is missing/invalid, if the input image doesn't exist, or if the script returns an error
Consider trimming the filename generation section — Claude can infer timestamp formatting from a single example rather than needing four examples plus detailed rules
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient but includes some redundancy — the usage section shows full command examples, then the Examples section at the bottom repeats very similar commands. The parameter mapping tables (e.g., 'quick/fast/draft -> low') and filename generation examples add useful value but the overall document could be tightened by ~20-30%. | 2 / 3 |
Actionability | Fully executable bash commands with all parameter options clearly specified. The commands are copy-paste ready with real paths, and the parameter mappings give Claude concrete decision rules for translating user intent into flags. | 3 / 3 |
Workflow Clarity | The skill is relatively single-task (run a script), but there's no validation or error handling guidance — no mention of what to do if the script fails, if the API key is invalid, if the image file doesn't exist, or if the output should be verified. For a skill that makes API calls and writes files, some error recovery guidance would be appropriate. | 2 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and headers, but it's a fairly long monolithic document (~100+ lines of content). The detailed parameter mappings, filename generation rules, and editing mode explanations could potentially be split into referenced files, though for a single-script skill this is borderline acceptable. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- intellectronica/agent-skills
- Commit
9b0e00a
- Reviewed
Table of Contents
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.