C

content-moderation

Siftr is a browser extension that reads your feed so you don't have to take it at face value. It analyzes social media content for rhetorical manipulation, misinformation signals, and AI-generated imagery, then returns an authoritative, cited response on what you're actually looking at. Three capabilities drive it: rhetoric analysis that identifies manipulation tactics like false urgency and emotional loading; misinformation signals that flag claims contradicting established sources or matching known archetypes; and local image authenticity analysis that detects AI generation artifacts without sending images off your device. When you want more than a flag, Siftr queries a verification API for a verdict with supporting citations and a concise explanation. Built on proven open-source components and modeled on feed-interception architecture, with a privacy-first posture: image analysis is fully local, text verification is opt-in, and there is no tracking, no analytics, and no data sales.

GitHub fork of ooguz/turkce-kufur-karaliste. Root upstream: ooguz/turkce-kufur-karaliste. License: CC-BY-SA-4.0. Original authors and contributor profiles: UPSTREAM_ATTRIBUTION.md

Official Python, TypeScript/Node, and MCP-server clients for the SentiSift comment-moderation and intelligence API. pip install sentisift · npm install @sentisift/client · uvx sentisift-mcp