Media.io AI Character Generator Skill#

Overview#

This skill calls Media.io OpenAPI to run character generation using model code character-generator-media-2.0. The API is asynchronous:

1. Submit generation request and get task_id.
2. Poll task result endpoint until the task is finished.

When To Use#

• The user wants stylized character images from prompt plus reference image.
• The user can provide an image URL reachable by Media.io servers.
• The user wants task-based generation with polling.

When Not To Use#

• The user asks for local file upload only (this API expects image URL input).
• The user asks for non-Media.io providers.
• The user asks for real-time synchronous image output in one call.

Requirements#

Environment Variables#

VariableRequiredDescriptionMEDIAIO_API_KEYYesMedia.io OpenAPI key, used in header X-API-KEY.

Base Headers#

• Content-Type: application/json
• X-API-KEY: $MEDIAIO_API_KEY

Supported Endpoints#

1) Query Credits#

• Method: POST
• Endpoint: https://openapi.media.io/user/credits
• Body: {}
• Purpose: check available credits before generation.

2) Create Character Task#

• Method: POST
• Endpoint: https://openapi.media.io/generation/effects/character-generator-media-2.0
• Body:

json

{
	"data": {
		"images": "https://example.com/input.jpg",
		"prompt": "anime style fantasy heroine",
		"ratio": "9:16"
	}
}

• Required fields:
  • data.images (string URL)
  • data.prompt (string)
• Optional fields:
  • data.ratio (string): 9:16, 16:9, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9

3) Query Task Result#

• Method: POST
• Endpoint: https://openapi.media.io/generation/result/{task_id}
• Body: {}
• Path parameter:
  • task_id (string, required)

Request and Response Contract#

Common Success Envelope#

json

{
	"code": 0,
	"msg": "",
	"data": {},
	"trace_id": "..."
}

Create Task Response#

• On success, data.task_id is returned.

Task Result Response#

• data.status can be one of the following values:

  • waiting: queued
  • processing: running
  • completed: completed successfully
  • failed: failed
  • timeout: timed out

• data.reason: provides additional context (e.g., success or error message)

• When status is completed:

  • data.result is an array of output objects with generated URLs
  • Each result object contains val (internal path), preview (public HTTPS URL), and status (completion status)

Standard Invocation Flow#

1. Call user/credits to verify balance.
2. Call character-generator-media-2.0 with data.images, data.prompt, and optional data.ratio.
3. Extract task_id.
4. Poll generation/result/{task_id} every 3 to 5 seconds.
5. Stop when status is completed or failed.
6. Return output URLs from data.result when completed.

cURL Examples#

Query Credits#

bash

curl --request POST \
	--url https://openapi.media.io/user/credits \
	--header 'Content-Type: application/json' \
	--header "X-API-KEY: $MEDIAIO_API_KEY" \
	--data '{}'

Create Character Task#

bash

curl --request POST \
	--url https://openapi.media.io/generation/effects/character-generator-media-2.0 \
	--header 'Content-Type: application/json' \
	--header "X-API-KEY: $MEDIAIO_API_KEY" \
	--data '{
	"data": {
		"images": "https://example.com/input.jpg",
		"prompt": "anime style fantasy heroine",
		"ratio": "9:16"
	}
}'

Query Task Result#

bash

curl --request POST \
	--url https://openapi.media.io/generation/result/<task_id> \
	--header 'Content-Type: application/json' \
	--header "X-API-KEY: $MEDIAIO_API_KEY" \
	--data '{}'

Successful Response Example

json

{
	"code": 0,
	"msg": "",
	"data": {
		"task_id": "effect-86f0f82a-36dc-4a7c-928a-721a18ef482f",
		"status": "completed",
		"reason": "success",
		"result": [
			{
				"val": "aicloudtmp/550160908/3/202603/1/combo_tm_alg-20260317165022-802800-60eb3-dwt.png",
				"preview": "https://url_to_generated_image.png",
				"status": "completed"
			}
		]
	},
	"trace_id": "a18315ba568b5c34407808d12cbc8457"
}

Response fields when status is completed:

• data.task_id: unique task identifier

• data.status: completed indicates successful completion

• data.reason: success indicates no error occurred

• data.result: array of output objects, each containing:

  • val: internal file path of the generated asset
  • preview: publicly accessible HTTPS URL for the generated asset
  • status: completed for each result item

Error Handling Guidance#

• Treat code != 0 as failure.
• Typical authentication errors:
  • 374004: not authenticated. Apply for an APP KEY at https://developer.media.io/.
• Typical request validation error:
  • 490000: params error
• Typical billing/credits error:
  • 490505: insufficient credits. Recharge before invoking generation APIs.
• Always include trace_id in logs for troubleshooting.

Agent Behavior Requirements#

• Validate that input contains a non-empty image URL and prompt before calling the create endpoint.
• Do not claim immediate output after task creation; always poll by task_id.
• If credits are insufficient, return a clear message and stop instead of retry loops.
• Avoid exposing raw API keys in logs or responses.

Safety and Compliance Notes#

• Only process user-provided or user-authorized images.
• Do not imply identity verification or biometric certainty from generated images.
• Generated output is synthetic media and should be presented as edited content.

References#

• Media.io platform: https://developer.media.io/
• API documentation: https://platform.media.io/docs/