$ ask steve’s bot > an ai guide to the work, projects and ideas of stefan preu, agentic systems architect. ask about pm cli or how he builds with agents. it can rewrite this page while you talk.
# your agent can ask too: mcp · a2a · cli
try
what is stefan working on? how does pm cli keep agent context? show me a project and the thinking behind it hack this page. surprise me. $ cat steve.txt # offline copy
stefan “steve” preu is an agentic systems architect in austria. he designs the systems around ai agents: what they know when a session starts, which tools they can use, where the boundaries are. his main project is pm cli, open-source project and context management for people and agents, with or without git (npm @unbrained/pm-cli). he also builds agent-friendly command-line tools for home assistant, jellyfin, searxng and monica, and publishes his open-source work as unbrained. ↗ about stefan · pm cli · unbrained.dev · contact stefan
# the chat needs javascript. privacy · imprint