Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ChristophLabestin/image-generator-mcp --skill image-generationgit clone --depth 1 https://github.com/ChristophLabestin/image-generator-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/christophlabestin/image-generator-mcp/image-generation)<a href="https://agentmods.dev/skills/christophlabestin/image-generator-mcp/image-generation"><img src="https://agentmods.dev/badge/skills/christophlabestin/image-generator-mcp/image-generation/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/christophlabestin/image-generator-mcp/image-generation"><img src="https://agentmods.dev/badge/skills/christophlabestin/image-generator-mcp/image-generation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00090 | $0.01351 |
| Opus 5 | $0.00045 | $0.00675 |
| Sonnet 5 | $0.00018 | $0.00270 |
| Haiku 4.5 | $0.00009 | $0.00135 |
Grade A, and why
image-generation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Image generation with GPT Image
Images are produced by the image-generator MCP server, which wraps OpenAI's
/v1/images endpoints. Three tools: generate_image, edit_image,
list_image_models.
Choosing a model
Leave model unset (or "auto") unless you have a reason not to — the server
starts at gpt-image-2 and falls back automatically if the key lacks access.
Override when cost or character matters:
| Situation | Model |
|---|---|
| Final artwork, text inside the image, 2K/4K, inpainting | gpt-image-2 |
| Lots of images, quality still matters, no 4K needed | gpt-image-1.5 |
| Cheap drafts, thumbnails, composition roughs | gpt-image-1-mini |
| User explicitly asks for DALL·E 3 | dall-e-3 |
If a call fails with a model error, run list_image_models — it reports which
models this API key can actually see — and retry with an available one.
The two-pass workflow
Generation is billed per image and high quality is several times the price of low. For anything non-trivial:
- Draft —
quality: "low", small size. Show the user, confirm the composition is right. - Final — same prompt,
quality: "high", the real size.
Skip the draft pass only for one-off throwaway images or when the user is clearly in a hurry.
Writing the prompt
These models reward long, concrete prose over keyword soup. Cover:
- Subject — what it is, in detail.
- Composition — framing, camera angle, what's in fore/background.
- Style/medium — photograph, 3D render, flat vector, watercolour, and so on. Name a lens and lighting for photographs.
- Palette and mood.
- Text — any words that must appear go in the prompt verbatim, in quotes. Say where they sit. Keep it short; long strings still come out garbled.
- Negatives — state them positively where you can ("an empty desk" beats "no clutter").
Do not silently rewrite what the user asked for. Add craft detail, keep intent.
Sizes and format
1024x1024square,1536x1024landscape,1024x1536portrait — the safe set for every model.gpt-image-2additionally takes arbitrary sizes: both edges multiples of 16, max edge 3840, aspect ratio under 3:1. Use it for1920x1080,2048x2048,3840x2160.background: "transparent"withoutput_format: "png"(orwebp) for logos, icons, stickers and anything that gets composited. Every gpt-image model supports this —gpt-image-2,1.5,1and1-miniwere all verified to return a real alpha channel with fully transparent corners. So you can draft transparent assets cheaply ongpt-image-1-miniand only go up for the final.dall-e-3cannot do transparency at all.output_format: "jpeg"for photographic images headed for the web.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 119 lines · 90 tokens per session scan A 298a8660cdfd
image-generation is a skill published in the GitHub repository ChristophLabestin/image-generator-mcp (0 stars, last pushed 14d ago), licensed MIT. It adds 90 tokens to every session and 1,351 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…