A useful annotation answers a specific question: where should the reader look, what should they notice, or what should they do next? One arrow and a short sentence are often enough. This guide explains how to add those details to an existing screenshot in the lightingshot image editor. Image editing and export happen locally in your browser; installing the capture extension is optional when you already have an image.

Choose a message before choosing a tool

Imagine reporting a checkout problem. A screenshot of the whole desktop leaves the reader to find the issue. A tighter image showing the checkout panel, an arrow pointing to the disabled button, and the label “Button stays disabled after all fields are complete” gives them a concrete starting point. Keep reproduction steps and longer explanations in the message accompanying the image.

Use an arrow for a single target, a rectangle for a group of controls, and a highlight for a short passage that must remain readable. Freehand drawing works for a quick circle or rough indication. Avoid placing every kind of mark on the same image; the marks compete with the content you want someone to understand.

Add your first annotation

  1. Open your screenshot. Open the editor and select 导入图片 (Import image) at the top. Choose a local image file. The sample picture is only a practice canvas; replace it before annotating your own work. Importing another image resets the current edits, so export anything you want to keep first.
  2. Frame the subject. If the screenshot includes unrelated windows, consider cropping it before adding marks. Leave enough surrounding interface to make the target understandable.
  3. Select an arrow. Click 箭头 (Arrow). Set 颜色 (Color) and 粗细 (Stroke width). Drag from an empty space toward the item you want to identify; the arrowhead appears at the end of the drag.
  4. Add a short label. Choose 文字 (Text), set 字号 (Font size), and click where the label should begin. Type in the dialog and select 添加文字 (Add text). You can use line breaks for two short lines. Position the label on a clear area so it does not cover the target.
  5. Review and export. Check the entire image, then click 导出图片 (Export image). Enter a useful filename, choose PNG, JPEG, or PDF, and select 下载文件 ↓ (Download file). Open the downloaded result before sending it.

Match the mark to the explanation

矩形 (Rectangle) draws an outline, making it useful for identifying a panel without hiding its contents. 高亮 (Highlight) places a translucent colored rectangle over an area. Choose a color with enough contrast and use it sparingly: several overlapping highlights can make small text hard to read. 画笔 (Pen) follows your pointer for freehand marks.

The color and width controls apply to marks you add afterward. They do not restyle an annotation already on the image. Existing text labels are not movable text boxes in this release. If an annotation lands in the wrong place, use 撤销 (Undo) and add it again. 重做 (Redo) restores an undone action until you make a new edit.

Make the exported image readable

The plus, minus, and 适合窗口 (Fit to window) controls change your editing view. They do not enlarge the exported pixels. A label that looks comfortable while zoomed in may look tiny when the recipient sees the whole image in a chat window. Preview the image at roughly the size the reader will see and shorten the label if necessary.

For small interface text and crisp marks, PNG is a useful starting choice. JPEG is another option when the image contains photography and file size matters. PDF packages the picture into A4 image pages; it does not turn the screenshot into searchable text. The export format guide explains those tradeoffs.

Check before sharing

Ask whether someone unfamiliar with the problem can identify the target immediately. Remove redundant marks, check that no label hides relevant evidence, and look for private information elsewhere in the picture. Highlighting does not hide information; use the separate opaque cover workflow when details need to be concealed in the exported image.

Keep the original separately if you need another version later. Downloads are image outputs rather than editable project files. If importing or downloading fails, follow the size limits and download checks on the help page.