Self-hosted AI coworkers, each with its own computer and a policy gateway
Emerged finds new products worth a look, daily — what’s new today
DeepSeek's open-source agent harness where every capability is a plugin
A major lab has open-sourced the scaffolding around its agents, so you can swap models, tools, sandboxes and UI as plugins instead of forking a monolith.
DeepSeek Harness (dsh) is an MIT-licensed agent harness from DeepSeek AI. Built on the Cordis kernel, which only loads, unloads and wires plugins, it treats models, tools, skills, sessions, sandboxes, storage, loops, scheduling and the UI as replaceable plugins. Run it with npx @deepseek-ai/dsh web for a local web UI. It is a developer preview with breaking changes expected, and a community plugin ecosystem is already forming.
Self-hosted AI coworkers, each with its own computer and a policy gateway
The one-click plugin and theme marketplace inside DeepSeek Harness
Durable AI agents you can self-host, on any model and any machine
DeepSeek's research build of an agent that watches one repo
Zhipu's open-source workbench for running AI agents on a codebase
An open-source coding agent built to cut what you spend on models