Skip to main content
Glama

mcp-media-engine

MCP server for MediaEngine — AI-powered image, audio, and video generation for Claude, Cursor, and other MCP clients.

Tools

Tool

Description

generate_image

Generate images from text prompts — full model catalog (Nano Banana, GPT-Image)

generate_article_hero_image

Hero banner for blog posts and articles

generate_email_header_image

Header image for marketing emails

generate_social_teaser_image

Social media graphics for LinkedIn, Twitter, Instagram

generate_video

AI video generation with provider-specific params

generate_article_teaser_video

Short teaser videos from article content

generate_audio

Text-to-speech synthesis

generate_article_narration

Audio narration for articles and blog posts

wait_for_job

Poll until a media job completes

get_job_status

Check status of a media job

list_jobs

List media jobs with filtering and pagination

Related MCP server: multimodal-mcp

Installation

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "mediaengine": {
      "command": "npx",
      "args": ["-y", "mcp-media-engine"],
      "env": {
        "MEDIAENGINE_BASE_URL": "https://your-mediaengine-instance.com",
        "MEDIAENGINE_API_KEY": "your-api-key"
      }
    }
  }
}

Cursor

Add to your .cursor/mcp.json:

{
  "mcpServers": {
    "mediaengine": {
      "command": "npx",
      "args": ["-y", "mcp-media-engine"],
      "env": {
        "MEDIAENGINE_BASE_URL": "https://your-mediaengine-instance.com",
        "MEDIAENGINE_API_KEY": "your-api-key"
      }
    }
  }
}

Configuration

Environment Variable

Required

Description

MEDIAENGINE_BASE_URL

Yes

Your MediaEngine API base URL

MEDIAENGINE_API_KEY

Yes

Your MediaEngine API key

LOG_LEVEL

No

Log level (debug, info, warn, error). Default: info

Quality Tiers

Tier

Model

Description

basic

Nano Banana

Fast and affordable

standard

Nano Banana

Balanced quality (default)

premium

GPT-Image

Highest quality — clean text rendering

License

MIT

Available Tools

11 tools
generate_article_hero_imageA

Generate a hero banner image for a blog post, article, or editorial page. Creates an AI-generated visual from the title and summary — ideal for CMS hero banners, newsletter headers, and article thumbnails. Use this when you need to create a featured image, hero graphic, or cover illustration for written content. Supports brand style hints and reference images.

Inputs: size, tone, title, summary, +5 more.

generate_article_narrationA

Generate audio narration of article or blog post content using text-to-speech. Creates podcast-style audio, accessibility voiceovers, and audio newsletters from written text. Pass the article title and body text; the tool selects an appropriate voice and submits an async audio job. Returns a job ID to poll for the finished audio file.

Inputs: text, speed, title, voice, +1 more.

generate_article_teaser_videoA

Create a short AI-generated teaser video (15–30s) from an article summary or social post. Produces a shareable video for driving traffic from video-first platforms like YouTube Shorts, TikTok, and Instagram Reels. Pass the article title and summary; the tool builds the generation parameters and submits an async video job. Supports tone and brand_style hints.

Inputs: tone, title, summary, provider, +4 more.

generate_audioA

Generate audio from text using text-to-speech (TTS) synthesis. Create voiceovers, narrations, podcast audio, or any spoken audio from text input. Supports multiple voices (alloy, echo, fable, onyx, nova, shimmer), formats (MP3, FLAC, AAC, Opus), and playback speeds. Use the intent-specific generate_article_narration tool for article content.

Inputs: text (required), speed, voice, format.

generate_email_header_imageA

Create an AI-generated header image for marketing emails and newsletters. Optimised for wide-format email banners (typically 600px wide) from subject line and body text. Accepts tone and brand_style hints to stay on-brand. Returns a queued media job whose result URL can be embedded in the email template. Use this to create email graphics, newsletter banners, or promotional header visuals.

Inputs: size, tone, title, summary, +5 more.

generate_imageA

Generate an image from a text prompt using the MediaEngine model catalog. Use this for any image creation request: illustrations, photos, AI art, logos, icons, graphics, or visuals. Accepts a raw prompt and optional aspect ratio, quality tier, model id, and advanced style parameters. Quality tiers: basic/standard (Nano Banana — fast, inexpensive default), premium (GPT-Image — strongest prompt adherence and clean text rendering). One integration, every model: new models are added to the catalog server-side, so the same wiring keeps working. Style guidance: put the desired look in the prompt itself (e.g. "black and white archival photograph", "flat vector illustration") and bake negatives into the prompt ("no text, no words, no logos"). Use the intent-specific tools (generate_article_hero_image, etc.) when you know the content type.

Inputs: prompt (required), size, model, quality, +13 more.

generate_social_teaser_imageA

Create a social media teaser image optimised for sharing on LinkedIn, Twitter/X, Instagram, or Substack. Auto-selects the best aspect ratio and generates a visual from article or post content. Accepts title, summary, tone, and brand_style. Use this to create social graphics, shareable images, post visuals, or og:image alternatives. Returns a queued media job.

Inputs: size, tone, title, summary, +5 more.

generate_videoA

Generate a video using AI video generation. Use this for any video creation request when the intent-specific video tools do not fit. Accepts provider-specific generation parameters directly.

Inputs: inputParams (required), provider.

get_job_statusA

Get the status and result of a media job by its ID.

Inputs: jobId (required).

list_jobsA

List media jobs with optional filtering, sorting, and cursor-based pagination.

Inputs: page, sort, type, limit, +6 more.

wait_for_jobA

Wait for a media generation job to complete — polls until the image, video, or audio is ready or the timeout is reached. Use this to get the final result without manually calling get_job_status in a loop. Returns the completed job on success or an error if the timeout expires.

Inputs: jobId (required), timeoutSeconds.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.1/5.0

    Scored across 11 tools

    Disambiguation4/5

    Tools are generally distinct with specific intent-based variants (e.g., generate_article_hero_image vs. generate_image). Overlap exists between general and specific generators, but descriptions clearly guide usage, reducing confusion.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern (generate_*, get_*, list_*, wait_for_*). No mixing of conventions, making the set predictable.

    Tool Count5/5

    11 tools cover the core media operations (generate images, audio, video in general and specific forms, plus job management) without being excessive. Scope is well-balanced.

    Completeness4/5

    The tool set covers generation and job lifecycle (status, listing, waiting). Missing delete/abort jobs is minor. Specific media types (article, email, social) are well-represented, leaving few gaps.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers