page_ocr
Extract text from any on-screen element or full viewport using local OCR. Read captchas, canvas text, and scanned UI without sending data to external services.
Instructions
TIER 2 VISION — LOCAL OCR: extract text from an element (or the whole viewport if no ref). 100% local (pure-Rust ML models auto-download once to GHOSTFOX_HOME/models). Answers 'what text is written there' — for image captchas, canvas text, scanned UI. Models download on first call (~12MB, once).
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"session_id": "<session_id>",
"page_id": "<page_id>"
},
"name": "page_ocr"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| ref | No | Ref of the element to OCR. Omit for the whole viewport. | |
| page_id | Yes | ||
| session_id | Yes |