AI 分析

核心用途是利用 macOS 系统原生的 Vision 框架进行本地图片文字识别。适合 macOS 用户，可在无网络或注重隐私的情况下为 Agent 提供图片解析能力。

包名

dsh-macos-vision-ocr

版本

1.0.0

许可证

MIT

Usage

Ask the agent to read an image, or call the tool explicitly:

{
  "file_path": "./scan.png",
  "languages": ["zh-Hans", "en-US"]
}

The result contains the canonical image path, recognized text, selected languages, and a truncated flag.

Edit this package's row in a later Harness patch layer when you need different defaults:

- id: dsh-macos-vision-ocr
  config:
    cacheDir: /absolute/path/to/cache
    languages: [en-US]
    maxOutputBytes: 2000000

Key	Default	Description
`cacheDir`	`$DSH_HOME/cache/ocr`	Swift source and compiled helper cache.
`languages`	`zh-Hans`, `zh-Hant`, `en-US`	Languages used when a tool call omits them.
`maxOutputBytes`	`1000000`	Maximum captured OCR stdout per call.