BlockShield for X

BlockShield for X Hide blocked, muted, and unwanted accounts across X feeds, replies, quotes, and search. Local-only privacy. --- BlockShield is a local-only content control layer for X.com. It lets you hide, blur, or label posts from accounts and keywords you choose — all within your own browser, with no data ever sent to any server. No servers. No login. No tracking. No account access. Your rules live in your browser, and nowhere else. This is privacy-first software by design. Why BlockShield? X's native block and mute features are powerful, but they don't always work as expected. Blocked accounts can still appear in quote tweets when other users quote them. They can show up in search results. They can appear in recommendations and "who to follow" widgets. Muted keywords can slip through algorithmic timelines and for-you feeds. When you block someone on X, that account can still be quoted into your timeline by someone else, and you see it — even though you explicitly blocked the original account. BlockShield exists to solve this gap. It restores your intent by enforcing your personal block and mute rules everywhere on X — your home feed, replies, quote tweets, search results, and recommendations. It acts as a local enforcement layer that works alongside X's own controls, not instead of them. Think of it as a second filter that catches what slips through. The result is a cleaner, calmer experience. You see only what you want to see, based on rules you control completely. How it works BlockShield runs entirely in your browser as a content script. When you visit X.com, it scans the page for tweet articles using stable DOM selectors that match X's data-testid attributes. For each tweet, it extracts the author handle, all mentioned handles, and the visible text. It then compares that data against your personal rule set — blocked handles, muted words, and muted phrases. If there is a match, it applies your chosen action: hide, blur, or label. The scanning is continuous. As you scroll and new tweets load via infinite scroll, a debounced MutationObserver detects the DOM changes and processes only new, unprocessed articles. This keeps performance smooth even during long scrolling sessions. The entire process is local and instantaneous — no network requests, no round trips, no delays. When you update your rules through the popup or options page, the page is rescanned immediately without requiring a refresh. The extension also detects URL changes when you navigate between different parts of X — home, search, profiles, tweet details — and rescans automatically. Features Block handles Hide posts from accounts you choose. Works in your home feed, tweet replies, quote tweets, and search results. Add handles manually in the popup or options page, or use the in-page shield button to block any author directly from their post. Handle input accepts multiple formats: @username, username, https://x.com/username. All formats are normalized to lowercase for consistent matching. Mute words Hide posts containing specific terms. Uses regex word-boundary matching, so "cat" matches standalone "cat" and "cats" but not "catalyst" or "catastrophe". This prevents false positives from partial word matches. All matching is case-insensitive. Minimum word length is two characters to avoid overly broad filtering. Mute phrases Hide posts containing exact phrases. "cat food" matches "I bought cat food today" and "CAT FOOD sale" but does not match when only "cat" or "food" appear separately. This gives you precise control over multi-word patterns. Case-insensitive. Three display modes Hide — completely removes matched posts from the DOM. Posts are hidden with display:none and retain no layout space. Blur — blurs the post content with an 8px Gaussian blur and reduces opacity. An overlay card appears with the match reason, a "Show once" button to temporarily reveal the post, and an "Allow @handle" button to add the account to your allowlist. Label — adds a visible orange warning banner above the post without hiding or bluring any content. Useful for moderation workflows where you want visibility into what is being flagged. Allowlist Exempt specific handles from all blocking and muting. If you have a blocked handle that you occasionally want to see, or a muted word that a specific account uses acceptably, add the handle to the allowlist. Allowlisted accounts bypass all rules, including word and phrase muting. In-page shield button A small shield icon appears at the top-right corner of each tweet when you hover over it. Clicking opens a contextual menu with options: Hide @author (adds to blocked handles), Allow @author (adds to allowlist and removes from blocked), Show this post once (temporarily reveals a blurred or hidden post), and Mute selected text (if you have text selected on the page, it offers to add it as a muted word). Quick popup controls Click the BlockShield icon in your Chrome toolbar to open a compact popup. From here you can toggle the entire extension on and off, switch between hide/blur/label modes, add a handle to block, add a word to mute, and see counts of your current blocked handles and muted items. A button opens the full options page. Full options page The options page organizes everything into tabs: Blocked Handles (with search and filter for long lists), Muted Words, Muted Phrases, Allowed Handles, Behavior Settings (toggles for filtering in quotes, replies, and search), and Import/Export. You can export your complete configuration as a JSON backup file, export your blocked handles as plain text, or import a previously saved JSON file. The danger zone at the bottom of Settings allows a complete factory reset with confirmation. Permissions BlockShield requests only the minimum permissions needed to function: storage — saves your blocked handles, muted words, muted phrases, allowed handles, and settings locally in your browser using Chrome's built-in storage.local API. https://x.com/* and https://twitter…