OCR for PowerPoint (.pptx) presentations with scanned or image-embedded slide content. Uses MinerU to extract text from image-based slides. Features: OCR ext...
Security Analysis
high confidenceThe skill's requirements, instructions, and install steps are consistent with an OCR CLI wrapper for MinerU; it needs the mineru-open-api binary and a MINERU_TOKEN, which matches the described behavior.
Name/description (PowerPoint OCR using MinerU) matches required binary (mineru-open-api) and required environment variable (MINERU_TOKEN). The declared install methods (npm package and Go install from the project's GitHub) are appropriate for providing the needed CLI.
SKILL.md only instructs the agent to call the mineru-open-api CLI and to set MINERU_TOKEN. This is within scope, but the workflow implies uploading or transmitting .pptx files to the MinerU service for remote OCR; users should be aware that files will be sent to an external service and that the token grants that service access. The instructions do not request unrelated files, env vars, or system paths.
Installers are standard language-specific installs: an npm package 'mineru-open-api' and a Go 'go install' from a GitHub repo. These are expected for a CLI distributed via those ecosystems. No downloads from untrusted personal servers or URL shorteners are specified.
Only one environment variable is required (MINERU_TOKEN), which is proportional for a remote API/CLI requiring authentication. There are no unrelated credentials or extra env vars requested.
Skill is not always-enabled and does not request any elevated platform privileges. It does not modify other skills or system-wide settings per the provided spec.
Guidance
This skill appears coherent, but before installing: 1) Understand that OCR likely uploads your .pptx files to MinerU's service — do not use it with sensitive/private slides unless you trust MinerU's privacy/retention policy. 2) Obtain MINERU_TOKEN only from https://mineru.net and verify the token's scope/permissions. 3) Prefer installing from the official GitHub/npm package names shown; inspect the npm package and GitHub repo if you can to confirm authenticity. 4) Test with non-sensitive sample files first. If you need on-device-only OCR, look for a local-only tool rather than a networked API.
Latest Release
v0.4.0
SEO: expand description for better ClawHub vector search discovery
More by @mzlzyca
Published by @mzlzyca on ClawHub
