The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes Python scripts (scripts/image_ocr.py and scripts/pdf_ocr.py) via the shell to perform OCR tasks. These scripts take user-provided file paths or URLs as arguments, which are passed to argparse and subsequently used in file operations or HTTP requests.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from images and PDF files.
Ingestion points: Data enters the agent's context from scripts/image_ocr.py (via result['text']) and potentially from downloaded OCR results via downUrl in scripts/pdf_ocr.py.
Boundary markers: There are no explicit instructions or delimiters in the SKILL.md to treat the OCR output as untrusted or to ignore any embedded instructions within the recognized text.
Capability inventory: The skill has file system access (via open()) and network access (via requests).
Sanitization: There is no evidence of sanitization, escaping, or filtering of the text extracted from documents before it is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill interacts with iFlytek's official API endpoints (cbm01.cn-huabei-1.xf-yun.com and iocr.xfyun.cn) to upload files and download OCR results. These are vendor-owned resources.

iflytek-pdf-image-ocr