openocr-skill
PassAudited by ClawScan on May 1, 2026.
Overview
This OCR skill is coherent with its stated purpose, but users should expect local file access, code execution, and possible model downloads to run OpenOCR.
This skill appears appropriate for local OCR and document parsing. Before installing, be aware that it may execute Python/OpenOCR workflows, read the images or PDFs you provide, and download model files if they are missing.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may run local OCR-related code and access files you provide for recognition or parsing.
The skill declares broad local tool access. This is expected for running OCR code and reading user-provided documents, but it is still worth noticing because those tools can affect local files or execute code if used outside the intended OCR workflow.
tools: [computer, code_execution, file_operations]
Use the skill only with files you intend to process, and review any proposed code or file operations before approving broader actions.
The skill may need to fetch OCR model files before processing documents, which can involve network access and external dependency trust.
The OpenOCR examples indicate that missing models may be downloaded automatically. This appears purpose-aligned for OCR model setup, but it means runtime behavior may depend on external model sources not fully captured by the registry metadata.
auto_download=True, # Auto-download missing models
Prefer trusted OpenOCR releases and model sources, and avoid running the skill in environments where unreviewed model downloads are not allowed.