Agnes Media MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AGNES_API_KEY | Yes | Your Agnes API key from https://www.agnes-ai.cn | |
| AGNES_OUTPUT_DIR | No | Absolute path to output directory for generated images and videos |
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
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| agnes_image_generateB | Generate an image with Agnes Image 2.1 Flash and save returned image data when possible. Supports text-to-image and image-to-image (via image_urls). Size accepts exact pixel dimensions such as '1024x768', '1024x1024', or '768x1024'. |
| agnes_image_generate_v2A | Generate an image with Agnes Image 2.1 Flash optimized for high-information-density visuals. Supports text-to-image and image-to-image (via image_urls). Size accepts tier values '1K', '2K', '3K', '4K' combined with ratio ('1:1', '3:4', '4:3', '16:9', '9:16', '2:3', '3:2', '21:9'). Legacy exact sizes like '1024x768' are also accepted. |
| agnes_image_editB | Edit or compose images through Agnes img2img using URLs or local files. Uses agnes-image-2.1-flash by default. Pass image_paths as public URLs or local file paths. Multiple images enable multi-image composition. |
| agnes_video_submitB | Submit an Agnes video task and return its video id for status polling. Supports text-to-video, image-to-video (via image URL), and keyframe animation (via extra_body with mode='keyframes'). num_frames is aligned to the 8n+1 rule automatically. |
| agnes_video_statusB | Fetch the current status for an Agnes video task by video_id. |
| agnes_video_waitB | Poll a video task by video_id until it completes, fails, or times out. |
| agnes_video_generateA | Submit an Agnes video task and wait for completion. Combines submit + wait into a single call. Returns the final video URL and optionally downloads the video file locally. |
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 7 tools
The image tools are ambiguous: agnes_image_generate and agnes_image_generate_v2 both generate images using the same model, with v2 merely adding tier-based sizes while also accepting legacy sizes, making it unclear which to use. Additionally, agnes_image_edit overlaps with generate's image-to-image mode, adding further confusion.
Tools mostly follow an agnes_<resource>_<action> pattern, but agnes_video_status is a noun phrase rather than an action (e.g., check_status), and agnes_image_generate_v2 introduces a version suffix that breaks the pattern. These minor deviations make the set readable but not fully consistent.
Seven tools is within the ideal range for a media generation server. While there is some redundancy (image v1 vs v2, video generate combining submit+wait), the overall count is well-scoped and not excessive.
The server covers core image generation, image editing, and video generation workflows (submit, status, wait, generate). Minor gaps exist such as lack of task listing or cancellation, but the primary generation and polling lifecycle is complete for typical usage.