Get Started
Link Whisper Premium icon

15 versions

Link Whisper Premium history.

Every release of Link Whisper Premium is archived here with its changelog, file size, and security scan result. Use the archive to roll back to a stable release or audit what changed between updates.

15 of 15 releases scanned clean
v4.0.5Latest
October 19, 20268.5 MBScanned
October 14, 20268.4 MBScanned
September 18, 20268.4 MBScanned
v4.0.2Unavailable
September 12, 20268.3 MBScanned
v4.0.1Unavailable
September 8, 20268.3 MBScanned
August 7, 20268.3 MBScanned
v2.9.5Unavailable
July 19, 20268.3 MBScanned
* [BUGFIX] Fixing loading error on PHP 8.3+ (It would show up as a "Headers already sent" error in the logs)
v2.9.3Unavailable
July 14, 20268.3 MBScanned
* [IMPROVEMENT] Improving the AI Linking's sentence relatedness checks to surface more valid links and skip unrelated ones. (We now pre-filter phrases and sentences before they hit the AI, verify that supplied words actually appear in the target post's content, and enforce sensible min/max word limits so the checker spends its effort where it counts) * [IMPROVEMENT] Improving how the Autolinking and AI Linking handle ignore words, including matching stemmed versions of your ignore words when cleaning and counting content. * [IMPROVEMENT] Skipping posts that already have active suggestions when building the AI link map, and preserving your data on a task reset instead of clearing it. * [EFFICIENCY] Adding a credit estimate for the AI sentence relatedness checks so the projected cost of an AI Linking run is more accurate before you start. * [EFFICIENCY] Filtering out and tags, and trimming unlikely-to-be-related sentences, before AI processing and making sure each sentence is only sent to the AI once to cut down on wasted processing and credits. * [COMPATIBILITY] Expanding the cache clearer to purge more caches, including SiteGround's OPcache and Memcache(d), so stale object cache entries don't linger after changes. * [COMPATIBILITY] Removing the temperature argument from OpenAI requests on newer models that no longer support it, so those models work correctly. * [COMPATIBILITY] Improving support for Rank Math redirects by sanitizing post slugs before they're used in the database query, fixing MySQL errors caused by malformed or reserved redirect slugs. * [BUGFIX] Fixing bugs that could cause broken links to drop off the report list. * [BUGFIX] Fixing an issue where selecting two links for the same post (same keyword and URL) would silently drop the second one, even with "Add link if post already has this link" turned on. * [BUGFIX] Fixing click tracking so logged-in editors are never tracked, regardless of whether visitor data collection is enabled. * [BUGFIX] Fixing the "Generate Sitemaps" button placement on the Visual Sitemaps report so the sticky top-right button and the centered button no longer collide. * [SECURITY] Sanitizing host and post slug inputs before they're used in queries.
v2.9.2Unavailable
June 17, 20268.2 MBScanned
* [NEW FEATURE] Adding an "Export Map" button to the AI Fix popup so you can download the relation map the fix is working with as a JSON file for review or troubleshooting. * [NEW FEATURE] Adding a "Rescan is_empty embedding posts" repair tool to the AI Settings that re-scans posts mistakenly saved as empty AI embeddings and then refreshes their AI Relation calculations. (For cases where an embedding request failed and good posts got incorrectly marked as empty) * [NEW DEBUG SETTING] Adding a "Clear OP Cache" tool to the Debug Settings so you can ask PHP to flush its OPcache and load freshly-updated plugin files. (Helpful when a host keeps serving an old cached version of the plugin and reports aren't reflecting recent changes) * [ADJUSTMENT] Sunsetting the direct OpenAI connection and migrating the plugin's AI features and on-screen messaging over to the Link Whisper AI connection. (The AI provider now defaults to the Link Whisper AI service, and the various "OpenAI" prompts and labels now reference the unified AI connection) * [IMPROVEMENT] Improving the AI credit estimator so it no longer adds estimated costs for processes that don't actually consume credits, making the One Click Setup / Dashboard scan estimates more accurate. * [IMPROVEMENT] Improving AI credit tracking so the credits used by the sentence-to-content relatedness comparison are now counted and logged in the credit history. * [IMPROVEMENT] Refreshing your available AI credit total whenever the purchase history is refreshed, so the displayed balance stays in sync after a purchase. * [IMPROVEMENT] Improving the text cleaning that runs before content is sent for embedding so it strips   characters and collapses runs of whitespace, producing cleaner, more accurate embeddings. * [IMPROVEMENT] Improving the AI Linking's relation map handling so a resumed run can re-process the existing map with a fresh AI pass instead of being stuck on the same stale map, and so the plugin no longer wastes time running OpenAI batch processing when the Link Whisper AI service is connected. * [IMPROVEMENT] Improving the AI Fix popup so its progress counter reflects the items actually available to work on once a run completes, and so stale saved suggestions are cleared when starting a fresh fix run. * [UI/UX] Formatting AI credit costs with thousands separators (e.g. "1,234 Credits") so estimated costs are easier to read. * [UI/UX] Rounding AI credit charges greater than 1 credit and hiding zero-credit rows in the credit history export so the report is clearer. * [BUGFIX] Fixing an accounting bug that made it look like every relation needed to be recalculated whenever a new post was published. * [BUGFIX] Fixing the basic Dashboard scan's completion check so a single straggling post no longer keeps the scan from being marked complete. * [BUGFIX] Fixing PHP notices/errors that could occur during AI relation processing when embedding data arrived as objects instead of arrays, and when the OpenAI streaming filter wasn't registered (the "empty index" notice). * [BUGFIX] Clearing the "Insufficient AI Credits" warning automatically once a refreshed balance shows you have credits available again.
v2.9.1Unavailable
June 9, 20268.2 MBScanned
* [NEW FEATURE] Creating a Dashboard Basic AI Scan flow! (AI scanning has been moved out of One Click Setup and into the Dashboard so the initial setup can focus on the site scan, while AI work can be handled from the Dashboard.) * [NEW FEATURE] Adding stalled and failed scan detection to the Dashboard AI scan. (If the scan stops checking in or runs into an error, Link Whisper now shows what happened and gives the user a retry path.) * [NEW FEATURE] Adding a "Process Now" option for Fix with AI features. (Now you can begin processing posts with AI based on a partial mapping of the site without waiting for the full mapping pass to finish.) * [IMPROVEMENT] Switching to decimal-based AI Credit reporting for greater accuracy. * [IMPROVEMENT] Improving AI Inbound Suggestions speed! (Inbound AI suggestions now prioritize the most related posts and avoid wasting AI work on low-value candidates.) * [IMPROVEMENT] Adding AI relation prechecking to filter out unrelated content before processing. * [IMPROVEMENT] Improving link insertion tracking. (Link insert actions better track whether insertion succeeded, including for links inserted into meta content.) * [UX/UI] Setting the Dashboard's Site Health widget to AJAX load. (The widget now loads site health data asynchronously so the Dashboard can open faster, particularly on large sites.) * [UX/UI] Adding a live Fix with AI post counter! (Large Fix with AI jobs now show how many posts have been analyzed and can estimate how much time is left.) * [UX/UI] Making Fix with AI planning more informative! (The mapping preview now updates with better progress, current estimates, and available post/link counts while planning is still running.) * [EFFICIENCY] Caching the Target Keyword relations for faster processing. * [EFFICIENCY] Setting the Inbound Suggestions to use the keyword engine if AI credits run out during processing. (Doesn't impact Fix with AI, only applies to the Inbound Suggestions) * [COMPATIBILITY] Improving ACF field handling during link insertion. * [COMPATIBILITY] Skipping posts hidden by redirects during AI processing. * [COMPATIBILITY] Guarding against empty AI relation data. (AI relation calculations now safely skip posts with empty or malformed relation indexes.) * [SECURITY] Improving AJAX-based feature security and improving the handling procedures for some report variables. * [BUGFIX] Fixing AI inbound suggestion retry loops! (Inbound AI suggestion generation no longer gets stuck retrying when the embedding data is empty.) * [BUGFIX] Improving AI link inserting checking to make sure links are inserted and following up when insertion needs another try. * [BUGFIX] Fixing URL changer and keyword URLs with encoded spaces! (Autolink/keyword URLs trim both regular spaces and trailing %20 padding before saving.) * [BUGFIX] Fixing minor datatype notices that could sometimes show up during suggestion generation.

Why archive

Sometimes the latest release isn't the one you need. The archive lets you pin Link Whisper Premium to a known-good version, or roll back while a bug is investigated. Every release stays scanned and reachable.