Oksskolten is a self-hosted, AI-integrated RSS reader designed to provide users with full-text access to articles by default, without requiring configuration or manual setup. It addresses the challenge of fragmented and ad-heavy online reading by automatically fetching, cleaning, and rendering articles as Markdown within its interface, eliminating the need to visit original sites or deal with intrusive layouts. The platform is suitable for those who want to streamline their news consumption, automate content extraction, and enrich articles with AI-powered tools.
The tool supports a wide range of sources, including RSS, Atom, and any standard web page, even if no feed is available. For sites without RSS, Oksskolten uses large language model (LLM)-powered selector inference to convert webpages into feeds, and employs FlareSolverr to handle JavaScript-rendered content and bot-protected sites. Its article fetching pipeline includes adaptive polling intervals, change detection, and safe SSRF-guarded retrieval. Content is processed through Readability and over 500 custom patterns to remove ads, navigation, and tracking elements, and is then converted to fenced-code GitHub Flavored Markdown. Extracted articles are stored in a WAL-mode SQLite database and indexed with Meilisearch for full-text, typo-tolerant search.
AI features are built in, allowing users to summarize, translate, and chat about articles directly in the app. Summaries provide concise overviews and bullet points, translations preserve Markdown formatting, and chat functionality maintains full conversation history. Users can choose from multiple LLM providers (such as Claude, Gemini, and OpenAI) and translation engines (including DeepL and Google Translate), supplying their own API keys. These AI capabilities are also available via an MCP server, enabling terminal-based assistants to interact with articles, so users can switch between web and CLI interfaces as needed.
Oksskolten is delivered as a single-container, self-hosted solution with no external database or subscription required. 0. Customization options include 14 color themes, 4 layout styles, and 9 fonts, with support for importing custom themes. The platform is positioned as a comprehensive, AI-native news reader for individuals seeking control and flexibility in their reading experience.
Oksskolten sits in PulseGate's News & reading category. It allows users to read, summarize, translate, and interact with articles from any feed or webpage using AI. It is built as an open-source project for news readers and information enthusiasts. Oksskolten is open source under the AGPL-3.0 license. Oksskolten is available on the web and the command line, and it can be self-hosted.
Oksskolten first shipped in 2026. Development happens publicly on GitHub with 422 stars and 150 commits in the last 90 days. Key capabilities include AI summarization, translation, and article chat. It exposes integrations via an MCP server.
Latest indexed changes and source events
Other apps tracked under the same category.