007.7: Screen OCR¶

• Status: Planned
• Date: YYYY-MM-DD
• Author(s): João V. Farias

Context¶

Provide context for capturing a screen region and extracting text directly to the Wayland clipboard.

Benchmark¶

Tesseract¶

• Shell pipeline: grim + slurp -> tesseract -> wl-copy (current workflow).

PaddleOCR / RapidOCR / Docling¶

Qwen-VL¶

Interaction Flow¶

Describe the interaction flow when triggering OCR from the launcher, selecting a screen region, copying recognized text to wl-copy, and notifying via 005: Notifications.

Visual Design¶

Provide visual design mockups or wireframes for the OCR launcher command and selection overlay.

Implementation¶

Describe the OCR script pipeline and notification feedback integration.