Lewati ke konten utama
Konten ini belum tersedia dalam bahasa Anda dan ditampilkan dalam bahasa Inggris.

Data Extractor

Skill Peringatan
85

>

Ringkasan AI

This skill leverages the unstructured Python library to process a wide range of document types, including PDFs, Word docs, emails, and HTML. It automatically detects and partitions elements, extracts text and metadata, and supports advanced features like table structure inference, OCR, and semantic chunking for RAG applications.

Scope

  • critical:Description qualityThe description is materially misleading as it contains only a single character ('>') and provides no actual information about the extension's functionality, which is contrary to the provided content in SKILL.md.

Documentation

  • info:Configuration & parameter referenceWhile the SKILL.md provides extensive code examples, it does not explicitly document all configuration options or parameters for the `partition` function or its variations, nor does it detail precedence order for any potential configurations.

Code Execution

  • info:ValidationThe SKILL.md demonstrates the use of `unstructured` library functions, which likely perform internal validation on file paths and parameters, but explicit schema validation within the skill's logic is not showcased.

Compliance

  • info:GDPRThe skill extracts data from documents. While it doesn't explicitly handle personal data, the extracted content could potentially contain PII, which would be submitted to the LLM without additional sanitization by this skill itself.

Instalasi

npx skills add claude-office-skills/skills

Menjalankan Vercel skills CLI (skills.sh) via npx — memerlukan Node.js secara lokal dan setidaknya satu agen yang kompatibel dengan skills terinstal (Claude Code, Cursor, Codex, …). Mengasumsikan repo mengikuti format agentskills.io.

3 months ago
98 stars
MIT
Diperbarui 3 days ago
Lihat sumber