PDF Text Extractor

The manifest and SKILL.md claim 'zero dependencies' and OCR via Tesseract.js, but package.json declares pdfjs-dist as a dependency and package-lock.json lists many packages; Tesseract.js is not present. The code dynamically requires 'pdfjs-dist' (it will error if not installed), so the 'zero dependencies' claim is false. OCR is advertised but index.js contains no Tesseract integration or OCR fallback — it always attempts text-layer extraction. These mismatches show the claimed capability does not align with the actual required components.

SKILL.md and README instruct using extractText with ocr:true and show examples assuming synchronous or fully-implemented OCR behavior. The runtime index.js does not implement OCR (no Tesseract usage) and also contains API misuse/bugs: extractText calls countWords(fullText) where countWords expects an object {text, options}, which will cause a runtime error; test.js treats extractText's Promise result as synchronous in places. The instructions therefore do not reflect what the code actually does and grant no clear guidance for installing required dependencies.

There is no install spec in the registry, but package.json and package-lock.json are included and declare pdfjs-dist and many nested packages. README suggests running 'npm install pdfjs-dist'. The absence of an install spec combined with packaged dependency manifests is inconsistent with the 'zero dependencies' marketing and means a user may need to run npm install (which pulls many packages and optional native build scripts). That increases friction and risk compared to the claimed zero-dependency design.

The skill does not request any environment variables, credentials, or config paths. Nothing in the files reads external secrets or unrelated system config.

Flags show the skill is not always-enabled and uses the default model-invocation behavior. It does not request persistent system-wide privileges or modify other skills' configuration.