document-to-image

分类: 内容与多媒体 | 上传者: ClassRobot | 下载: 0 | 版本: v1.0（最新）

当需要在 OCR、多模式理解或预览渲染之前将 Word、PPT 或 PDF 文件规范化为图像时，请使用此技能。

更新日志: Source: GitHub https://github.com/ClassRobot/ClassRobot

目录结构

当前层级: tree/new/skills/document-to-image/

📄 runtime.py 88 B
📄 SKILL.md 954 B

SKILL.md

---
name: document-to-image
description: Use this skill when Word, PPT, or PDF files need to be normalized into images before OCR, multimodal understanding, or preview rendering.
---

# Document To Image

Use this skill when the input is a `.doc`, `.docx`, `.ppt`, `.pptx`, or `.pdf` file and downstream logic works better on page images than on the original binary document.

## Workflow

1. Prefer `utils.skills.document_to_image_skill.convert(...)` when the source may be bytes, local paths, or remote URLs.
2. Prefer `utils.skills.document_to_image_skill.convert_local_file(...)` when the source is already a local file and you need explicit output paths.
3. Reuse `utils.tools.docs2img` as the low-level converter implementation instead of duplicating Office/PDF conversion logic.

## Notes

- This skill is the canonical entry for document-to-image normalization inside the project.
- The runtime implementation lives in `utils/skills/runtime.py`.

登录后下载/点赞/收藏 ❤ 9 | ★ 0

document-to-image

目录结构

SKILL.md

举报内容

提示