Build media generation into your AI agent workflows.
MCP Media Engine is an async media generation API for the Model Context Protocol. It lets agents like Claude generate images, video, and audio — no design tools, no API plumbing.
Quickstart
Get an API key
Sign up at mcpmediaengine.com and copy your API key from the dashboard. Test keys (prefixed me_test_) return placeholder images instantly.
Add to Claude Desktop
Add to claude_desktop_config.json: { "command": "npx", "args": ["-y", "mcp-media-engine"], "env": { "MEDIAENGINE_API_KEY": "me_live_your_key" } }
Ask Claude to generate
Start a conversation and say: "Generate a hero image for my blog post about [topic]." Claude calls the MCP tool and returns the image URL when ready.
MCP Tool Reference
MCP Media Engine exposes these tools via the Model Context Protocol. Agents call them directly — no custom integration code required.
generate_article_hero_imageGenerate a hero banner image for a blog article. Claude picks up the title and summary automatically and selects an appropriate visual style.
params: title, summary?, tone?, brand_style?
generate_social_teaser_imageGenerate a social media teaser image. Auto-selects the correct aspect ratio for LinkedIn, Twitter, Facebook, or Instagram.
params: title, summary?, target_platform, brand_style?
generate_email_header_imageGenerate an email newsletter header image, optimised for 600px-wide email templates.
params: title, summary?, brand_style?
generate_imageGeneric text-to-image generation. Specify any model (astria, dall-e-3, stable-diffusion) and size directly.
params: prompt, model?, size?
generate_videoGenerate a short-form video clip from a text description. Long-running job — results delivered via webhook or polling.
params: prompt, duration?, model?
generate_audioGenerate speech or voiceover audio from text. Choose from six voices and set output format.
params: text, voice?, format?
get_job_statusPoll for the status of a previously submitted job. Returns status, asset URL (if complete), or error details.
params: job_id
list_jobsList recent media generation jobs with optional filtering by status, media type, and date range.
params: status?, media_type?, limit?, cursor?
Full tool reference: github.com/modelcontextprotocol/servers
How to generate a blog hero image with Claude
MCP Media Engine is an MCP-native API that lets Claude generate images for blog posts automatically. Claude detects the article topic and calls generate_article_hero_image, returning the asset URL when ready.
This works with any MCP-compatible client (Claude Desktop, Cursor, Windsurf) and any Paperclip agent — no custom code, no API plumbing.
What is the Model Context Protocol?
MCP is an open standard that lets AI agents connect to external tools and data sources. MCP Media Engine implements this standard, so any MCP-compatible agent can access image, video, and audio generation without custom integration code.
How long does image generation take?
Typically 10–30 seconds for images. Video generation is longer (minutes). Both are processed asynchronously — your agent submits a job and receives the result via polling or webhook. Long-running jobs never block the request thread.
How do I receive the generated image?
Claude automatically polls get_job_status until the image is ready and returns the asset URL in the conversation. Alternatively, pass a webhook URL when generating to receive the result via HTTP POST.
Example: Claude Desktop
Minimal Claude Desktop config for MCP Media Engine
{
"mcpServers": {
"mediaengine": {
"command": "npx",
"args": ["-y", "mcp-media-engine"],
"env": {
"MEDIAENGINE_API_KEY":
"me_live_your_key_here"
}
}
}
}Then in Claude:
"Generate a hero image for my post about AI productivity tools for writers."
Quickstart
Get an API key and generate your first image in 5 minutes.
API Reference
Full endpoint reference for the REST API and MCP tools.
Pricing
25 free credits to start. No credit card required.