SafeShot — Full Element Screenshot

About

SafeShot — Full Element Screenshot

SafeShot captures what no built-in screenshot tool can: the complete content of any scrollable element or an entire web page — stitched into a single, pixel-perfect image or exported as a clean multi-page PDF in standard A4 format.

Whether you need to document a long chat thread, archive a full news article, save a product listing, capture an analytics dashboard, or preserve a legal webpage for your records, SafeShot handles it in seconds with two clicks.

HOW IT WORKS

Click the SafeShot icon in your toolbar to open the control panel. You have two capture modes:

► Pick Element
Click "Pick Element", then hover over any part of the page. SafeShot highlights elements as you move your mouse. Click the one you want — a chat pane, a scrollable table, a long feed, a sidebar — and SafeShot automatically scrolls through it, stitching every frame into one seamless image. Press Escape at any time to cancel.

► Full Page
Click "Full Page" to capture the entire document from top to bottom, regardless of how long it is. No element selection needed — SafeShot scrolls the whole page and assembles every viewport into a single output.

Once the capture is complete, choose your export format:

→ PNG — Downloads the raw stitched image immediately. Full resolution, lossless, ready to attach to an email, paste into a document, or share directly.

→ PDF — Exports a multi-page A4 PDF. Each page is sized to the A4 format with the image scaled to fill the full page width. Every page includes a compact header showing the page title, URL, and page number, and a footer showing the capture timestamp with timezone. Add optional custom footer text — a project name, classification label, or your organisation's name — using the Footer Text field in the popup.

THREE INTELLIGENT CAPTURE MODES

SafeShot automatically detects the best capture strategy for each situation:

Mode A — Scroll Container
When you pick an element that itself scrolls (a chat panel, an overflow container, a custom list), SafeShot scrolls the element internally and crops each frame to that element's exact boundaries. The rest of the page is never included.

Mode B — Page Scroll
When you pick an element that is taller than the viewport but relies on the window to scroll (a long article, a full-width table, a tall card), SafeShot scrolls the window and crops every frame to the element's bounding rectangle.

Mode C — Full Page
When you use Full Page mode, SafeShot scrolls the entire document from top to bottom. Fixed and sticky elements — navigation bars, cookie banners, floating headers — are automatically hidden from frame 1 onwards so they don't repeat throughout the stitched image (frame 0 preserves the header so it appears naturally at the top of the output).

HANDLES TRICKY PAGES AUTOMATICALLY

Fixed and sticky elements handled correctly
Navigation bars, floating headers, cookie consent banners, and sticky sidebars are automatically detected and hidden during multi-frame captures. SafeShot even pierces Shadow DOM, catching fixed elements inside web components (like YouTube's navigation bar). They are fully restored after capture — the page is left exactly as you found it.

Non-scrollable pages work correctly
Some pages report a larger scroll height than they actually scroll (due to overflow:hidden, absolute positioning, or browser quirks). SafeShot detects when the scroll position stops advancing and stops automatically, trimming the output canvas to the content that was actually captured.

High-DPI and Retina displays supported
Every pixel calculation accounts for devicePixelRatio. Screenshots taken on a 2× or 3× display are handled correctly — no tearing, no misalignment, no blurry stitching.

Rate limiting handled internally
Chrome limits screenshot API calls to approximately two per second. SafeShot enforces this limit with an internal timer and retries automatically if the limit is briefly exceeded, so you never see an error from a fast capture.

Long captures protected by a watchdog
If a capture stalls for more than five minutes (for example because the tab was closed or navigated away), SafeShot automatically heals the state and shows an error rather than leaving the popup stuck indefinitely.

PDF EXPORT — BUILT FOR PROFESSIONAL USE

The PDF export is designed to produce documents you can hand to a colleague, attach to a report, or file for compliance purposes.

A4 format, always
Every page is standard A4 (210 × 297 mm / 595 × 842 pt). The captured image is scaled to fill the full A4 width, and page breaks are computed from that scale so the aspect ratio is always correct. No custom page sizes, no awkward margins.

Compact, readable header on every page
Each page carries a single-line header containing the page title (left), the URL (centre, clipped to never overlap the page number), and the page number (right). Header height is just 20 pt — enough information, nothing wasted.

Timestamped footer with timezone
The footer on every page shows the exact capture date and time including the UTC offset — for example: 22.04.2026 - 14:35:10 (UTC+02:00). This makes it unambiguous when and where the capture was taken, which matters for compliance documentation, legal records, and audit trails.

Optional custom footer text
Type anything into the Footer Text field — a project code, a classification label ("Confidential", "Draft", "Internal Only"), a department name, a client reference — and it appears right-aligned in the footer on every page.

Filenames include date, domain, and title
Downloaded files are named automatically: YYYY-MM-DD - domain.com Page Title.pdf (or .png). Filesystem-unsafe characters are stripped and the title is capped at 60 characters so filenames are always clean and sortable.

PRIVACY AND PERMISSIONS

SafeShot requests only the permissions it actually uses:

• tabs — required to query the active tab ID, send messages to the content script, and call Chrome's screenshot API.
• storage — required to persist your Footer Text setting between sessions.
• host_permissions (<all_urls>) — required so the screenshot API can capture any tab regardless of its URL.

SafeShot does not collect, transmit, or store any data outside your browser. All captured images are processed entirely in memory within the extension and are never sent to any server. The stitched image is held in the background service worker's memory and discarded when you reset or close the extension. Nothing is uploaded anywhere.

SafeShot has no analytics, no telemetry, no sign-in, no account, no subscription, and no network requests of any kind beyond loading the Google Fonts stylesheet in the popup UI.

WHAT SAFESHOT IS GREAT FOR

✔ Capturing long chat conversations (Slack, Teams, WhatsApp Web, Gmail threads)
✔ Archiving full news articles or blog posts
✔ Documenting product listings on e-commerce sites
✔ Saving analytics dashboards and reports
✔ Capturing scrollable data tables and spreadsheets
✔ Creating evidence screenshots for legal or compliance purposes
✔ Preserving web pages that may change or disappear
✔ Producing A4 PDF reports from web content
✔ Documenting UI/UX designs and prototypes
✔ Capturing social media timelines and profiles
✔ Recording before/after states for testing or QA

TIPS

• To capture a specific scrollable element (e.g. a chat panel inside a larger page), use Pick Element and click directly on that panel — not the surrounding page.

• To capture the full document including everything below the fold, use Full Page.

• If you want to add a label to every page of your PDF (such as "Confidential" or a project name), type it into the Footer Text box before clicking PDF. It saves automatically between sessions.

• The PNG export downloads the raw capture with no modifications — ideal when you need the image for further editing.

• The PDF export processes the page breaks to fit A4 — ideal when you need a document you can print or file.

• If a capture seems stuck, clicking the button (which becomes "Cancel" during capture) will abort and reset to the ready state.

TECHNICAL DETAILS

SafeShot is a Manifest V3 Chrome extension with no external dependencies beyond the vendored jsPDF library for PDF generation. It has no build step and no runtime dependencies on external services.

The capture pipeline works as follows: the content script scrolls the target element or window in steps, sends a capture request to the background service worker after each step, receives the screenshot data URL, and draws each frame onto an offscreen canvas at the correct destination Y coordinate. A minimum 550 ms gap between successive screenshot calls is enforced to comply with Chrome's rate limit. When all frames are collected, the canvas is converted to a PNG data URL and held in the service worker's memory (not written to storage, which has a 5 MB quota that large captures can exceed). The popup retrieves the data URL on demand when you click PNG or PDF.

Supported: Chrome 109 and later (Manifest V3 service workers with the tabs and storage APIs).