How these extensions capture screenshots
Chrome gives extensions a method to capture the visible part of a tab as an image. A documentation extension calls it each time you click, type or change page, and pairs the image with a description of the element you touched: its label, its role, where it sits on the page. That description is what turns a folder of screenshots into a guide.
Two limits follow from how this works. The capture shows the visible viewport, not the whole scrolling page. And Chrome does not let extensions read or capture certain pages, including chrome:// pages and the Chrome Web Store.
Questions to ask before installing
- Where do the screenshots go? Some tools upload every capture to the vendor's servers. Others keep them in the browser. For internal tools, this is the question that matters.
- What permissions does it want? “Read and change all your data on all websites” is the broadest. An extension that only acts when you press a button can ask for much less.
- What does it capture in password fields? The answer should be nothing, not blurred afterwards.
- Can you blur before you share? And is that free?
- Can you take your data with you? Look for open formats: Markdown, HTML, PDF, DOCX.
What Showsteps asks for, and why
| Permission | Why |
|---|---|
| activeTab | Lets you start a recording on the current tab with one click, and lets Showsteps screenshot browser-internal pages that extensions cannot otherwise capture. |
| scripting | Injects the recorder into the pages you visit while a recording runs, and removes it when you press Stop. |
| storage | Keeps your settings and the in-progress recording state if Chrome suspends the background worker. |
| sidePanel | Shows the step list and editor beside the page. |
| unlimitedStorage | Guides with many screenshots are large; this stops the browser evicting them. |
| Site access (optional) | Requested only when you press Record, so it can screenshot and notice clicks in every tab you visit. You can revoke it in Chrome at any time. |
The privacy page has the full explanation. Showsteps has no analytics, no account and no network requests after install.
Sensitive data: what is automatic and what is not
Showsteps marks password, card and similar fields as sensitive while recording. Their values are never stored, and the field is masked before the screenshot is stored. It cannot know that a name in a table or an email in a header belongs to a real person, so it lets you draw a blur or crop on any screenshot before you export. Read each image before you send the file.
What a recording session looks like
You press Record, and a small indicator stays on the page so you always know it is running. Every click, selection, key such as Enter and page change adds a step. Typing is grouped, so filling in an email address makes one step rather than one per key. If you open a link in a new tab, or a sign-in redirect changes the site, the recording follows you.
You can pause while you deal with something that should not be in the guide, such as a message from a colleague, and resume afterwards. When you press Stop the steps appear in the editor, where you can delete a stray click before anything is exported.
Try any recorder on a throwaway task first, with made-up data, and export the result. That tells you in a few minutes whether the screenshots are readable, whether the step text needs much editing and whether the file opens where you need it.
Getting the documents out
You can export Markdown (a folder with guide.md and images), a self-contained HTML file, PDF, DOCX, an agent skill, or the .showsteps project itself, which is a zip with a guide.json and PNG images. Because the format is documented, you or a script can open, read and rewrite it without the extension.