# Feature request https://docs.jami.net/developer/feature-requests.html#sfl > * Public page/personal blog available through Jami (Contact #Jami / medium difficulty & long term project) ## Platform All: Linux, macOS, Windows ## Site hosting with Jami 1. Host (web)site on Jami. 1. Start with: * Static (web)site * HTML support * CSS support * Image `<img>` tag support 2. Another ticket can be created later to add email hosting. 2. Host (web)site on the device of the JamiID. * No domain name registration required. The site is accessible by visiting the JamiID with Jami. * No website hosting required on a third-party computer (or server). Site is stored on the users device. * Site would be stored in a folder under the JamiID, allowing for 1 site per JamiID. 3. Visit site * If the JamiID is unreachable to a contact, then the (web)site would be unavailable to the contact. * Clicking the **Visit site** would open the site with the default web browser installed on the contact's device. ## DHTNet https://git.jami.net/savoirfairelinux/dhtnet > The DHTNet library is designed to establish secure peer-to-peer connections using public-key authentication. Are changes required to DHTNet? ## Screenshots ### Conversations panel 1. In the **Conversations** panel, right-click on a conversation to show the contextual menu. 2. Add a `Visit site` button.  ### Selected conversation 1. Click on a conversation. 2. In the top-right, add a `Visit site` button.  ### Contact details Add a `Visit site` button in the contact details. What do you think? Thank you ### It's Not Just Claude: Most Top AI Models Will Also Blackmail You to Survive After Claude Opus 4 resorted to blackmail to avoid being shut down, Anthropic tested other models, including GPT 4.1, and found the same behavior (and sometimes worse).<br> By Michael Kan June 21, 2025<br> https://au.pcmag.com/ai/111675/its-not-just-claude-most-top-ai-models-will-also-blackmail-you-to-survive > As AI adoption continues to grow, maybe it’s best to avoid giving a chatbot access to your entire email inbox. > “We deliberately created scenarios that presented models with no other way to achieve their goals, and found that models consistently chose harm over failure,” Anthropic noted.