AI 产品解读Highlights links to people, pages, and communities on social networks with color-coded labels indicating whether they are trans-friendly or transphobic/anti-LGBT
It is difficult to tell at a glance whether a feminist or LGBT-related social media page/user is trans-friendly or transphobic. This extension provides crowd-sourced visual indicators so users can quickly assess content they encounter.
User installs extension → extension automatically loads bloom filter data → as user browses supported social networks, links to users/pages/communities are highlighted with trans-friendly (green) or transphobic (red) colors → user can right-click unlabeled links to contribute votes → votes are encrypted and submitted to the community API for aggregation into future bloom filter updates.
Color-codes links on Facebook, Reddit, Twitter/X, YouTube, Medium, Tumblr, Wikipedia, Bluesky, Mastodon instances, and search engines (Google, Bing, DuckDuckGo) as trans-friendly (green) or transphobic (red)Uses bloom filters (bundled + dynamically updated from GitHub) to store and look up known labels for millions of identifiersAllows users to right-click links and vote/mark them as 'anti-trans', 't-friendly', or clear labels via context menuSubmits encrypted user votes to a community API (shini-api.xyz) for aggregationSupports customizable color themes (green-red, purple-yellow, cyan-orange)Runs on 50+ social media sites including numerous Mastodon instances
- 目标用户
- LGBT community members and allies who want to quickly identify trans-friendly vs transphobic content online / Social media users browsing Reddit, Twitter, Facebook, YouTube, Wikipedia, Bluesky, and Mastodon
- The PayPal detection in the deterministic heuristic was a false positive — it appeared in a blocklist of unsupported social networks, not as actual payment integration
- Cannot verify the exact content of the binary bloom filter data files (t-friendly.dat, transphobic.dat) to confirm what identifiers are labeled
