Studio Image Generation

Use this skill when the user wants to generate an image or edit an existing image.

Always call Ekko Studio's media endpoint. Do not call an upstream image API directly, and do not ask the user for an API key. The server reads the selected/requested profile's config.yaml and uses a configured custom provider. By default it uses the provider named fun-codex, but callers may request another configured provider by sending provider, provider_name, or custom_provider.

This skill is separate from Hermes Agent's native image_generate tool. The native tool reads image_gen from config.yaml; this Studio-managed endpoint reads the auxiliary.image_generation and auxiliary.image_edit routes below. Changing one does not change the other.

Do not use any built-in image generation tool as a fallback. If the Hermes Web UI endpoint returns 401, 403, connection failure, or any other error, stop and report the Hermes Web UI error to the user.

custom_providers:
  - name: fun-codex
    base_url: https://api.apikey.fan/v1
    api_key: ...
    model: gpt-5.5
    api_mode: codex_responses

Example with another configured provider:

custom_providers:
  - name: agnes
    base_url: https://agnes.example/v1
    api_key_env: AGNES_API_KEY
    model: agnes-image-2.1-flash

auxiliary:
  image_generation:
    provider: agnes
    model: gpt-image-2
    timeout: 600
  image_edit:
    provider: agnes
    model: gpt-5.4-mini
    timeout: 600

image_generation supplies the route for text-to-image and multipart image edits. image_edit supplies the route and primary model for image-to-image through the Responses API. If the active route has no provider, the server falls back to the other image route and then to fun-codex.

Endpoint:

POST <Hermes Web UI base URL>/api/studio/media/apikey-image-generate

Resolve the Hermes Web UI base URL in this order: