ZenBird: Toxicity Filter for X (Twitter)

About

Keep your X (Twitter) feed peaceful, productive, and clutter-free with ZenBird.

Social media should be a space for connection, inspiration, and healthy discussions. But far too often, scrolling through X (Twitter) exposes you to unwanted negativity, toxic insults, political outrage, dog whistles, and targeted harassment.

ZenBird is a premium, lightweight, and ultra-high-performance content filtering extension that acts as a real-time shield for your mental health. It automatically scans, analyzes, and collapses toxic posts on your feed before they ever reach your eyes, replacing them with a polite, non-intrusive placeholder.

🛡️ CORE FEATURES (Included for Free)

Curated 2,700+ Word Default Dictionary: Instantly block known profanity, racial slurs, modern toxic slang, dehumanizing language, self-harm instigations, and evasive spelling variations (e.g., f*ck, s.h.i.t, k_y_s).
Custom User Blocks: Want to avoid specific topics, political figures, handles, or trending topics? Easily add your own custom words, hashtags, phrases, or handles directly from the popup dashboard.
Zero-Latency Native Processing: Designed with extreme performance in mind. Utilizing pre-compiled single-pass regular expressions, ZenBird scans hundreds of tweets in microseconds without inducing any browser lag or battery drain.
Seamless X (Twitter) Integration: Works perfectly with both x.com and twitter.com. The extension matches all background themes dynamically (Default, Dim, and Lights Out) to maintain X’s original visual aesthetic.
Real-Time Block Counter: View your digital wellness stats directly on your dashboard. ZenBird keeps track of exactly how many toxic posts have been safely shielded during your sessions.

🧠 UNLEASH THE POWER OF ZENBIRD PRO (Advanced AI Shield)

Standard keyword filters are great, but bad actors often find ways to bypass them. ZenBird Pro upgrades your extension with our state-of-the-art Smart AI Shield for advanced, context-aware comprehension:

Context-Aware Content Analysis: No more accidental mutes. Our AI reads the entire context of a post. It distinguishes between actual harassment and casual slang, ensuring you only block what is truly toxic.
Bullying & Sarcasm Detection: Standard word filters cannot read tone. ZenBird Pro detects subtle passive aggression, mockery, condescending sarcasm, and emotional abuse with high accuracy.
Zero-Latency Smart Bypassing: ZenBird Pro performs a local dictionary pre-scan first. If a post is clean or matches local rules, it bypasses the cloud API instantly—saving RAM, CPU, and zeroing latency.
Session-Smart Caching: Once a post has been analyzed, its toxicity evaluation is securely cached inside your active session. Scrolling back and forth incurs zero redundant network requests or overhead.
Cryptographically Signed Handshakes: Outbound requests to our classification backend are cryptographically signed using secure HMAC-SHA256 signatures, ensuring complete data security and tamper-proofing.

🎨 CHOOSE YOUR AESTHETIC

ZenBird is as beautiful as it is powerful. Customize your placeholder card theme to fit your mood or match your environment:

Auto (Match Twitter): Seamlessly blends into your light, dim, or pitch-black dark mode feed.
Classic Light / Dim / Black: Fixed styles to always keep elements rendered to your precise preferences.
Vibrant Gradients: Elevate your workspace with premium Light or Dark Glassmorphism gradients.

🔒 PRIVACY-FIRST ENGINEERING

We believe your digital behavior belongs to you alone.

No Data Logging: Standard filtering runs entirely locally on your device.
Anonymous AI Queries: When Pro AI is enabled, only the raw text of the post is securely sent to our classification servers. No personal details, IP logs, usernames, or account handles are ever associated with requests.
No Tracking: We do not sell user data, utilize ad trackers, or log your browsing history.