MeiGen AI Design MCP is an MCP server that lets AI coding tools generate images and videos through cloud models, OpenAI-compatible APIs, or local ComfyUI, an interface for running image-generation workflows. It is for creating designs such as logos, product images, animations, and batches of visual variations from coding-agent environments. Its catalogue entries provide skills, commands, agents, a plugin, and a hook for using the server's design workflow.
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 jau123/MeiGen-AI-Design-MCP --skill product-photoshootgit clone --depth 1 https://github.com/jau123/MeiGen-AI-Design-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/jau123/meigen-ai-design-mcp/product-photoshoot)<a href="https://agentmods.dev/skills/jau123/meigen-ai-design-mcp/product-photoshoot"><img src="https://agentmods.dev/badge/skills/jau123/meigen-ai-design-mcp/product-photoshoot/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/jau123/meigen-ai-design-mcp/product-photoshoot"><img src="https://agentmods.dev/badge/skills/jau123/meigen-ai-design-mcp/product-photoshoot.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00112 | $0.01200 |
| Opus 5 | $0.00056 | $0.00600 |
| Sonnet 5 | $0.00022 | $0.00240 |
| Haiku 4.5 | $0.00011 | $0.00120 |
Grade A, and why
Product Photoshoot Workflow 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 13d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Product Photoshoot Workflow
Turn one product reference photo into a brand-ready set of 4 distinct images, each emphasizing a different sales angle.
When to trigger
- User uploads a product photo and says "make e-commerce images", "design a campaign", "I need product shots"
- User says 电商产品图 / 产品多角度图 / 产品拍摄 / 产品营销图
- Anyone asking for "a set of product images" with a reference attached
Prerequisites
- A reference image is required — the user MUST provide a product photo (URL or local path). If they have not, ask once: "Please share the product photo you want me to work from." Do NOT try to invent a product without a reference.
- Confirm provider is configured (any of MeiGen / OpenAI-compatible / ComfyUI). If not, hand off to
/meigen:setup.
The 4 directions
Always plan exactly 4 directions in this order. Present as a table to the user and ask which to generate(per UX Rule 8 — always confirm before batch). Include "all four in parallel" as an option.
| # | Direction | Aspect | Intent |
|---|---|---|---|
| 1 | Lifestyle Scene | 4:3 or 16:9 | Product placed in its natural use-context (a watch on a wrist, a candle on a coffee table, a bottle on a bar). Soft natural lighting. |
| 2 | Macro Detail | 1:1 | Extreme close-up. Material grain, texture, surface reflections. Studio lighting. |
| 3 | Scale / Context | 4:3 | Product alongside a familiar object (hand, fruit, ruler) to convey size, OR product on a clean pedestal with subtle shadow. |
| 4 | Marketing Layout | 3:4 or 9:16 | Editorial / poster composition — product offset to one side, with negative space for headline text. Strong color block or gradient background. |
Generation flow
- Read user intent — product type, brand mood (luxury? playful? minimalist?), color palette. Use
search_gallery(category="Product & Brand")if you need style references; 239 curated examples are available. - Plan and present — write the 4 prompts (distinct, not just one tweaked four ways) and show them to the user. Each prompt MUST:
- Reference the input image's product
- Specify the direction's intent (lifestyle / macro / scale / marketing)
- Include lighting direction
- Mention material handling if relevant (matte, glossy, translucent, metallic)
- AskUserQuestion — "Pick directions to generate. All four runs ~4 parallel agents — confirm?"
- Generate in parallel — for the directions chosen, spawn
meigen:image-generatoragents in a single response, each passing the reference image asreferenceImagesand the planned prompt. OmitaspectRatiounless the user pinned one — describe the ratio in the prompt instead so each direction gets its own framing. - Present results — Image URL + saved path for each, grouped by direction label. Do NOT describe the generated images (UX Rule 1).
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.
- 13d ago First seen · 78 lines · 112 tokens per session scan A c5da3f540e24
Product Photoshoot Workflow is a skill published in the GitHub repository jau123/MeiGen-AI-Design-MCP (1,749 stars, last pushed 1mo ago), licensed MIT. It adds 112 tokens to every session and 1,200 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
sfcc-sfra-scss
Best practices for styling and theming SFRA storefronts using SCSS. Use when asked to create style overrides, theming, responsive layouts, or CSS customizations in SFCC.
sfcc-ocapi-hooks
Guide for implementing OCAPI hooks in Salesforce B2C Commerce. Use this when asked to create OCAPI hooks, extend API endpoints, validate API requests, or modify API responses.
sfcc-page-designer
Guide for creating Page Designer pages and components in Salesforce B2C Commerce.
sfcc-scapi-custom-endpoints
Guide for developing SCAPI Custom APIs on Salesforce B2C Commerce. Use this when asked to create custom REST endpoints, api.json, schema.yaml, or script implementations.
sfcc-caching
Unified caching playbook for SFCC (page cache vs custom cache vs service response cache). Use this when improving performance, reducing external calls, designing cache keys/TTLs, or debugging stale cache behavior.
sfcc-platform-limits
Cheat-sheet and design patterns for surviving SFCC quotas and limits (script timeouts, HTTPClient call caps, session size, custom object quotas, file I/O restrictions, and headless rate limits). Use this when debugging enforced quota violations or designing scalable SFCC architectures.