media-gen
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FAL_KEY | No | Optional API key that enables photorealistic image generation via FLUX and video generation via Kling. | |
| GEMINI_API_KEY | No | Optional API key that enables photorealistic image generation via Imagen 4 and video generation via Veo 3. | |
| OPENAI_API_KEY | No | Optional API key that enables photorealistic image generation via gpt-image-1 and video generation via Sora. | |
| MEDIA_OUTPUT_DIR | No | The directory where generated media files are saved. Defaults to 'generated-media/'. | generated-media/ |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| render_imageA | Render SVG markup or a full HTML document (authored by you) to an image file on disk. Use for illustrations, diagrams, charts, UI mockups, typographic cards. Not photorealistic. |
| render_videoA | Render an MP4 from JavaScript canvas animation code you author. Provide draw(ctx, t, frame) as a pure function of time t in seconds (deterministic — no requestAnimationFrame, no Date.now). Optional setup(ctx) runs once. Use for motion graphics, animated diagrams, explainers. Returns file path plus sample frames. |
| list_providersA | List configured BYOK generation providers, their models and capabilities. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
The three tools have distinct purposes: render_image produces static images, render_video produces animated video, and list_providers enumerates backend providers. There is no overlap or ambiguity between them.
Tool names follow a consistent snake_case verb_noun convention: render_image, render_video, list_providers. The verb precisely indicates the action and the noun indicates the output or resource, making the pattern predictable.
With exactly three tools, this server is tightly scoped to its media generation purpose and carries no redundant tools. Each tool earns a clear place in the set.
The server covers the apparent core workflow: discover available providers, render static visual assets, and render animated video assets. Since rendering is a stateless generate-to-file operation, no update/delete/list-generated-files tools are required for the declared scope.