RecognizeTextTool

Extracts text from a local image with on-device ML Kit text recognition (Latin script). Returns the full recognized text plus per-line text and bounding boxes.

No permissions required; image_path is sandboxed to the external-storage root via PathValidator and must point at an existing file. Read-only.

Output keys: text, line_count, lines (each with text and nullable bounding_box {left, top, right, bottom}).

Constructors

Link copied to clipboard
constructor(context: Context)

Properties

Link copied to clipboard
open override val annotations: ToolAnnotations
Link copied to clipboard
open override val description: String
Link copied to clipboard
open override val name: String
Link copied to clipboard
open override val parameters: List<ToolParameter>

Functions

Link copied to clipboard
open suspend override fun execute(params: Map<String, Any>): ToolResult