본문으로 건너뛰기
이 콘텐츠는 아직 사용자의 언어로 제공되지 않아 영어로 표시됩니다.

Data Extractor

Skill 경고
85

>

AI 요약

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.

설치

npx skills add claude-office-skills/skills

Vercel skills CLI(skills.sh)를 npx로 실행합니다. 로컬에 Node.js와 skills 호환 에이전트(Claude Code, Cursor, Codex 등) 중 하나 이상이 설치되어 있어야 합니다. 저장소가 agentskills.io 형식을 따른다고 가정합니다.

3 months ago
98 stars
MIT
1 day ago에 업데이트됨
소스 코드 보기