HangarVision guide

What it is

HangarVision turns screenshots and PDFs into context your agent can read. The fastest way: copy a screenshot and paste it into a Claude bay — the image is saved and its path is typed into the terminal for you. For more control, the full panel lets you annotate, then scan an image or PDF into text. Scans are a paid feature.

Before you start

• Hangarvibe.
• For a full scan, your own agent CLI installed (for example claude) — HangarVision runs the extraction through it.

How to use it

1. 1

   For a quick add, take a screenshot with Win+Shift+S.

2. 2

   Click inside a Claude bay and press Ctrl+V. The image is saved and its path is typed in for the agent. Dropping an image file onto a bay does the same.

3. 3

   For more control, open the full panel with Ctrl+Shift+V.

4. 4

   Drop, paste, or browse to a screenshot or PDF.

5. 5

   Optional: annotate it. Highlight the regions that matter, or crop the image.

6. 6

   Click Scan. Extraction runs through your own installed CLI, like claude, and takes about 20 seconds.

7. 7

   Inject the extracted text into a bay you choose, or copy it.

Good to know

• The fast copy-paste path is best for screenshots. The panel is best for PDFs, or when you want to annotate first.
• Everything is logged on your machine, so you can see what was scanned.