@deepseek-ai/dsh-desktop 404s on npm, the last five releases in the official repo have zero attachments, and searching the whole site for “download” turns up nothing. Follow those three leads and you conclude: there is no official desktop app.
awesome-dsh-plugin.com has a search box at the top with 4287 / 4287 in the corner. Below it, a row of category tags: UI 731, tools and capabilities 567, development and runtime 312, sessions and messaging 281, workflow 260, usage and billing 226, models 206, memory 202…
dsh is a web app: run npx to start the server, it takes port 3080, open a browser tab and work with the agent there. It works, but switch windows or close the browser and you have to go find it again. It always feels like living in someone else’s house.
Python’s agent frameworks have already cycled through several generations — LangChain, LlamaIndex, and now Google’s own ADK. Kotlin has been quiet the whole time — not because nobody wanted to build one, but because there was nothing official to lean on, and nobody quite dared put it in production.
While cleaning up his disk, tech blogger ferstar noticed the ~/.zcode directory was taking up a lot of space – over 700MB.
Last week Inc Magazine contributing editor Jason Aten posted some screenshots on Threads. He was chatting with Meta’s new AI assistant Muse, and partway through something felt off — Muse started asking him questions about conversations in his Messages.
Starting with yesterday’s news: Claude Code 2.1.277 now supports AGENTS.md (September 18).
One detail first, and it may strike you as a little strange.
¥65 million — the winning bid on a China Zheshang Bank agent infrastructure project. Add the wealth-management and supply-chain agent projects already in production, and Zheshang’s public agent procurement tops ¥74 million.
Anyone who writes Kotlin has been carrying the same nagging feeling for the past two years.