The latest articles on DEV Community by 0xzerolight (@0xzerolight). https://dev.to/0xzerolight https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4064757%2F63bfa506-6632-4329-8502-a7c6b77625d3.png https://dev.to/0xzerolight en 0xzerolight Wed, 05 Aug 2026 20:52:42 +0000 https://dev.to/0xzerolight/topic-watch-open-source-self-hosted-news-monitor-for-productivity-8jl https://dev.to/0xzerolight/topic-watch-open-source-self-hosted-news-monitor-for-productivity-8jl <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Ffdm0unte31smtgjvh9c8.gif" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Ffdm0unte31smtgjvh9c8.gif" alt="Topic Watch demo" width="720" height="382"></a></p> <h2> Topic Watch </h2> <p>Topic Watch is a self-hosted news monitor that only notifies you on truly new information using LLM-powered novelty detection.<br> Fully free and open-source.</p> <h2> Why not RSS readers? </h2> <p>RSS news readers notify you on every single new article on a topic. </p> <p>This often forces you to re-read the same exact story rehashed 20 times in the periods between truly new information.</p> <p><a href="https://github.com/0xzerolight/topic_watch" rel="noopener noreferrer">Topic Watch</a> tracks a topic and notifies you only when something is genuinely new.</p> <p>Thus, you save precious time while staying informed on the latest news.</p> <h2> Pipeline </h2> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>RSS / Bing News / Exa search ↓ fetch → dedup → extract content ↓ knowledge state + new articles → LLM ↓ has_new_info? ├─ no → record, stay silent └─ yes → update state → notify → record </code></pre> </div> <h2> Knowledge state, not keyword matching </h2> <p>Each topic has a knowledge state: a rolling LLM-written summary of what is already established. Every check compares the new articles against that summary to check for truly new details.</p> <p>Nothing new: the check is recorded, nothing is sent. <br> Something new: the state absorbs the fact, then a notification goes out with the summary, key facts, and source links.</p> <h2> Cost </h2> <p>You can use any LLM API via LiteLLM - Claude, ChatGPT, Grok, Deepseek, etc. The provider is a single config string.</p> <p>Roughly $0.0003 per check on a small model (e.g. Claude Haiku 4.5).<br> Or run Ollama locally and pay nothing. </p> <h2> GitHub link </h2> <p><strong>Fully self-hosted and open-source.<br> Feel free to share issues and feature ideas!<br> A star on GitHub is appreciated if you like it :).</strong></p> <p>Check it out:<br> <strong><a href="https://github.com/0xzerolight/topic_watch" rel="noopener noreferrer">github.com/0xzerolight/topic_watch</a></strong></p> opensource python productivity ai