Skill: image-analyze

Version: 1.0.0

Deep image analysis returning scene graph with objects, relations, and bounding boxes.

name: image-analyze version: 1.0.0 description: Analyze image content with objects, relations, and scene structure

Credentials

credentials:

• name: OPENAI_API_KEY required: false description: OpenAI API key

Skill inputs

inputs:

• name: image_url type: string required: true description: URL of the image to analyze
• name: detail type: string default: high enum: [low, high, auto] description: Analysis detail level

Skill outputs

outputs:

• name: analysis type: object description: "{objects, colors, text, scene_type, composition}"
• name: metadata type: object description: Processing metadata

Backend options

backends:

• name: openai type: remote requires: [OPENAI_API_KEY]
• name: ollama type: local models: [llava, bakllava]

Expanso components used

components: inputs: - stdin - http_server processors: - mapping - openai_chat_completion - log outputs: - stdout - sync_response