dsh-maclens

Bridge Apple's on-device Vision framework (macOS) into DeepSeek Harness (dsh) as local tools: OCR, image classification, face detection, and document layout.

• 🔒 100% local — no network, no API key, no daemon, no upload. The Vision framework is already on your Mac.
• ⚡ Fast — sub-second OCR for typical screenshots, running on the Neural Engine.
• 🇨🇳 Great Chinese OCR — Vision's recognition supports zh-Hans + 30+ languages out of the box.

Tools

Tool	What it does
maclens_ocr	Transcribe all visible text with confidence + bounding boxes
maclens_classify	Top-N category identifiers (document, chart, diagram, photo, …)
maclens_faces	Face bounding boxes and confidence
maclens_document	OCR + simple left/right column layout analysis

Install

Prerequisites: macOS (Vision framework), Xcode Command Line Tools (Swift).

# 1. clone or copy this repo, then build the Swift bridge
cd dsh-maclens
bash scripts/build.sh          # → bin/maclens

# 2. install the plugin into a dsh profile
dsh plugin --profile desktop add ./dsh-maclens     # local path install
# or, once published: dsh plugin --profile desktop add dsh-maclens

The plugin spawns bin/maclens (or swift/.build/release/MaclensBridge, or a maclens on PATH — set MACLENS_BIN to pin a specific binary).

How it works

dsh (text-only model)
  └─ maclens_ocr / classify / faces / document   ← dsh tools (lib/index.js)
       └─ bin/maclens (Swift CLI)                ← swift/Sources/MaclensBridge
            └─ Apple Vision framework (VNRecognizeTextRequest, VNClassifyImageRequest,
               VNDetectFaceRectanglesRequest)    ← on-device, offline

Development

cd swift && swift build -c release
.build/release/MaclensBridge ocr --image /path/to/img.png
.build/release/MaclensBridge classify --image /path/to/img.png --top 3

License

MIT